#install.packages('TDAmapper')
library(TDAmapper)
library(cluster)
library(vip)
##
## Attaching package: 'vip'
## The following object is masked from 'package:utils':
##
## vi
#install.packages('kernlab’)
library(kernlab)
#install.packages(‘class’)
library(class)
#install.packages('nnet')
library(nnet)
#install.packages(‘randomForest’)
library(randomForest)
## randomForest 4.7-1.1
## Type rfNews() to see new features/changes/bug fixes.
#install.packages('e1071')
library(e1071)
#install.packages("BayesFactor")
library(BayesFactor)
## Loading required package: coda
##
## Attaching package: 'coda'
## The following object is masked from 'package:kernlab':
##
## nvar
## Loading required package: Matrix
## ************
## Welcome to BayesFactor 0.9.12-4.5. If you have questions, please contact Richard Morey (richarddmorey@gmail.com).
##
## Type BFManual() to open the manual.
## ************
library(BayesPPD)
library(bayestestR)
#install.packages('igraph')
library('igraph')
## Warning: package 'igraph' was built under R version 4.3.3
##
## Attaching package: 'igraph'
## The following object is masked from 'package:BayesFactor':
##
## compare
## The following object is masked from 'package:class':
##
## knn
## The following objects are masked from 'package:stats':
##
## decompose, spectrum
## The following object is masked from 'package:base':
##
## union
#install.packages('locfit')
library(locfit)
## locfit 1.5-9.8 2023-06-11
#install.packages('ggplot2’)
library(ggplot2)
##
## Attaching package: 'ggplot2'
## The following object is masked from 'package:randomForest':
##
## margin
## The following object is masked from 'package:kernlab':
##
## alpha
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:igraph':
##
## as_data_frame, groups, union
## The following object is masked from 'package:randomForest':
##
## combine
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
#install.packages('networkD3')
library(networkD3)
library(rstanarm)
## Loading required package: Rcpp
## This is rstanarm version 2.26.1
## - See https://mc-stan.org/rstanarm/articles/priors for changes to default priors!
## - Default priors may change, so it's safest to specify priors, even if equivalent to the defaults.
## - For execution on a local, multicore CPU with excess RAM we recommend calling
## options(mc.cores = parallel::detectCores())
library(see)
#install.packages('tidyverse')
library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ lubridate 1.9.3 ✔ tibble 3.2.1
## ✔ purrr 1.0.2 ✔ tidyr 1.3.0
## ✔ readr 2.1.4
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ lubridate::%--%() masks igraph::%--%()
## ✖ ggplot2::alpha() masks kernlab::alpha()
## ✖ tibble::as_data_frame() masks dplyr::as_data_frame(), igraph::as_data_frame()
## ✖ dplyr::combine() masks randomForest::combine()
## ✖ purrr::compose() masks igraph::compose()
## ✖ purrr::cross() masks kernlab::cross()
## ✖ tidyr::crossing() masks igraph::crossing()
## ✖ tidyr::expand() masks Matrix::expand()
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ✖ ggplot2::margin() masks randomForest::margin()
## ✖ purrr::none() masks locfit::none()
## ✖ tidyr::pack() masks Matrix::pack()
## ✖ purrr::simplify() masks igraph::simplify()
## ✖ tidyr::unpack() masks Matrix::unpack()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
#install.packages('caret')
library(caret)
## Loading required package: lattice
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:purrr':
##
## lift
##
## The following objects are masked from 'package:rstanarm':
##
## compare_models, R2
#install.packages('ISLR')
library(ISLR)
#install.packages('MCMCpack')
library(MCMCpack)
## Loading required package: MASS
##
## Attaching package: 'MASS'
##
## The following object is masked from 'package:dplyr':
##
## select
##
## ##
## ## Markov Chain Monte Carlo Package (MCMCpack)
## ## Copyright (C) 2003-2025 Andrew D. Martin, Kevin M. Quinn, and Jong Hee Park
## ##
## ## Support provided by the U.S. National Science Foundation
## ## (Grants SES-0350646 and SES-0350613)
## ##
#linstall.packages("caret")
library(caret)
library(TDA)
##
## Attaching package: 'TDA'
##
## The following object is masked from 'package:cluster':
##
## silhouette
library(TDAstats)
library(ks)
##
## Attaching package: 'ks'
##
## The following object is masked from 'package:TDA':
##
## kde
##
## The following object is masked from 'package:MCMCpack':
##
## vech
##
## The following object is masked from 'package:igraph':
##
## compare
##
## The following object is masked from 'package:BayesFactor':
##
## compare
#install.packages('MLmetrics')
library(MLmetrics)
##
## Attaching package: 'MLmetrics'
##
## The following objects are masked from 'package:caret':
##
## MAE, RMSE
##
## The following object is masked from 'package:base':
##
## Recall
#install.packages('googledrive')
library(googledrive)
#install.packages('stringr')
library(stringr)
#install.packages('ks')
library(ks)
library(GGally)
## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2
#import DryBean dataset from UCI repository stored on my desktop
#Dry_Bean_Dataset **
library(readxl)
Dry_Bean_Dataset <- read_excel("~/Desktop/NCU/DissertationDatasets/DryBeanDataset/Dry_Bean_Dataset.xlsx")
head(str(Dry_Bean_Dataset))
## tibble [13,611 × 17] (S3: tbl_df/tbl/data.frame)
## $ Area : num [1:13611] 28395 28734 29380 30008 30140 ...
## $ Perimeter : num [1:13611] 610 638 624 646 620 ...
## $ MajorAxisLength: num [1:13611] 208 201 213 211 202 ...
## $ MinorAxisLength: num [1:13611] 174 183 176 183 190 ...
## $ AspectRation : num [1:13611] 1.2 1.1 1.21 1.15 1.06 ...
## $ Eccentricity : num [1:13611] 0.55 0.412 0.563 0.499 0.334 ...
## $ ConvexArea : num [1:13611] 28715 29172 29690 30724 30417 ...
## $ EquivDiameter : num [1:13611] 190 191 193 195 196 ...
## $ Extent : num [1:13611] 0.764 0.784 0.778 0.783 0.773 ...
## $ Solidity : num [1:13611] 0.989 0.985 0.99 0.977 0.991 ...
## $ roundness : num [1:13611] 0.958 0.887 0.948 0.904 0.985 ...
## $ Compactness : num [1:13611] 0.913 0.954 0.909 0.928 0.971 ...
## $ ShapeFactor1 : num [1:13611] 0.00733 0.00698 0.00724 0.00702 0.0067 ...
## $ ShapeFactor2 : num [1:13611] 0.00315 0.00356 0.00305 0.00321 0.00366 ...
## $ ShapeFactor3 : num [1:13611] 0.834 0.91 0.826 0.862 0.942 ...
## $ ShapeFactor4 : num [1:13611] 0.999 0.998 0.999 0.994 0.999 ...
## $ Class : chr [1:13611] "SEKER" "SEKER" "SEKER" "SEKER" ...
## NULL
summary(Dry_Bean_Dataset)
## Area Perimeter MajorAxisLength MinorAxisLength
## Min. : 20420 Min. : 524.7 Min. :183.6 Min. :122.5
## 1st Qu.: 36328 1st Qu.: 703.5 1st Qu.:253.3 1st Qu.:175.8
## Median : 44652 Median : 794.9 Median :296.9 Median :192.4
## Mean : 53048 Mean : 855.3 Mean :320.1 Mean :202.3
## 3rd Qu.: 61332 3rd Qu.: 977.2 3rd Qu.:376.5 3rd Qu.:217.0
## Max. :254616 Max. :1985.4 Max. :738.9 Max. :460.2
## AspectRation Eccentricity ConvexArea EquivDiameter
## Min. :1.025 Min. :0.2190 Min. : 20684 Min. :161.2
## 1st Qu.:1.432 1st Qu.:0.7159 1st Qu.: 36714 1st Qu.:215.1
## Median :1.551 Median :0.7644 Median : 45178 Median :238.4
## Mean :1.583 Mean :0.7509 Mean : 53768 Mean :253.1
## 3rd Qu.:1.707 3rd Qu.:0.8105 3rd Qu.: 62294 3rd Qu.:279.4
## Max. :2.430 Max. :0.9114 Max. :263261 Max. :569.4
## Extent Solidity roundness Compactness
## Min. :0.5553 Min. :0.9192 Min. :0.4896 Min. :0.6406
## 1st Qu.:0.7186 1st Qu.:0.9857 1st Qu.:0.8321 1st Qu.:0.7625
## Median :0.7599 Median :0.9883 Median :0.8832 Median :0.8013
## Mean :0.7497 Mean :0.9871 Mean :0.8733 Mean :0.7999
## 3rd Qu.:0.7869 3rd Qu.:0.9900 3rd Qu.:0.9169 3rd Qu.:0.8343
## Max. :0.8662 Max. :0.9947 Max. :0.9907 Max. :0.9873
## ShapeFactor1 ShapeFactor2 ShapeFactor3 ShapeFactor4
## Min. :0.002778 Min. :0.0005642 Min. :0.4103 Min. :0.9477
## 1st Qu.:0.005900 1st Qu.:0.0011535 1st Qu.:0.5814 1st Qu.:0.9937
## Median :0.006645 Median :0.0016935 Median :0.6420 Median :0.9964
## Mean :0.006564 Mean :0.0017159 Mean :0.6436 Mean :0.9951
## 3rd Qu.:0.007271 3rd Qu.:0.0021703 3rd Qu.:0.6960 3rd Qu.:0.9979
## Max. :0.010451 Max. :0.0036650 Max. :0.9748 Max. :0.9997
## Class
## Length:13611
## Class :character
## Mode :character
##
##
##
ggpairs(Dry_Bean_Dataset, aes(color = Class))
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggpairs(Dry_Bean_Dataset, columns = c(1:8,17), aes(color = Class))
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggpairs(Dry_Bean_Dataset, columns = c(9:17), aes(color = Class))
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
##Add Bayesian tests functions
#create function to conduct the Bayesian Sign Test
BayesianSignTest <- function(diffVector,rope_min,rope_max) {
library(MCMCpack)
samples <- 3000
#build the vector 0.5 1 1 ....... 1
weights <- c(0.5,rep(1,length(diffVector)))
#add the fake first observation in 0
diffVector <- c (0, diffVector)
#for the moment we implement the sign test. Signedrank will follows
probLeft <- mean (diffVector < rope_min)
probRope <- mean (diffVector > rope_min & diffVector < rope_max)
probRight <- mean (diffVector > rope_max)
results = list ("probLeft"=probLeft, "probRope"=probRope,
"probRight"=probRight)
return (results)
}
##Create function to conduct Bayesian Signed Rank Test
BayesianSignedRank <- function(diffVector,rope_min,rope_max) {
library(MCMCpack)
samples <- 30000
#build the vector 0.5 1 1 ....... 1
weights <- c(0.5,rep(1,length(diffVector)))
#add the fake first observation in 0
diffVector <- c (0, diffVector)
sampledWeights <- rdirichlet(samples,weights)
winLeft <- vector(length = samples)
winRope <- vector(length = samples)
winRight <- vector(length = samples)
for (rep in 1:samples){
currentWeights <- sampledWeights[rep,]
for (i in 1:length(currentWeights)){
for (j in 1:length(currentWeights)){
product= currentWeights[i] * currentWeights[j]
if (diffVector[i]+diffVector[j] > (2*rope_max) ) {
winRight[rep] <- winRight[rep] + product
}
else if (diffVector[i]+diffVector[j] > (2*rope_min) ) {
winRope[rep] <- winRope[rep] + product
}
else {
winLeft[rep] <- winLeft[rep] + product
}
}
}
maxWins=max(winRight[rep],winRope[rep],winLeft[rep])
winners = (winRight[rep]==maxWins)*1 + (winRope[rep]==maxWins)*1 + (winLeft[rep]==maxWins)*1
winRight[rep] <- (winRight[rep]==maxWins)*1/winners
winRope[rep] <- (winRope[rep]==maxWins)*1/winners
winLeft[rep] <- (winLeft[rep]==maxWins)*1/winners
}
results = list ("winLeft"=mean(winLeft), "winRope"=mean(winRope),
"winRight"=mean(winRight) )
return (results)
}
#Create function to conduct the Bayesian Correlated t.test
#diff_a_b is a vector of differences between the two classifiers, on each fold of cross-validation.
#If you have done 10 runs of 10-folds cross-validation, you have 100 results for each classifier.
#You should have run cross-validation on the same folds for the two classifiers.
#Then diff_a_b is the difference fold-by-fold.
#rho is the correlation of the cross-validation results: 1/(number of folds)
#rope_min and rope_max are the lower and the upper bound of the rope
correlatedBayesianTtest <- function(diff_a_b,rho,rope_min,rope_max){
if (rope_max < rope_min){
stop("rope_max should be larger than rope_min")
}
delta <- mean(diff_a_b)
n <- length(diff_a_b)
df <- n-1
stdX <- sd(diff_a_b)
sp <- sd(diff_a_b)*sqrt(1/n + rho/(1-rho))
p.left <- pt((rope_min - delta)/sp, df)
p.rope <- pt((rope_max - delta)/sp, df)-p.left
results <- list('left'=p.left,'rope'=p.rope,'right'=1-p.left-p.rope)
return (results)
}
set.seed(16974)
###Prepare drybean dataset for One hot encoding if necessary and Persistent homology.
##One hot encoding for drybean dataset
library(caret)
#define one-hot encoding function
dummy_drybean <- dummyVars(" ~ .", data=Dry_Bean_Dataset)
#perform one-hot encoding on data frame
dry_bean_dataset_one_hot_df <- data.frame(predict(dummy_drybean, newdata=Dry_Bean_Dataset))
summary(dry_bean_dataset_one_hot_df)
## Area Perimeter MajorAxisLength MinorAxisLength
## Min. : 20420 Min. : 524.7 Min. :183.6 Min. :122.5
## 1st Qu.: 36328 1st Qu.: 703.5 1st Qu.:253.3 1st Qu.:175.8
## Median : 44652 Median : 794.9 Median :296.9 Median :192.4
## Mean : 53048 Mean : 855.3 Mean :320.1 Mean :202.3
## 3rd Qu.: 61332 3rd Qu.: 977.2 3rd Qu.:376.5 3rd Qu.:217.0
## Max. :254616 Max. :1985.4 Max. :738.9 Max. :460.2
## AspectRation Eccentricity ConvexArea EquivDiameter
## Min. :1.025 Min. :0.2190 Min. : 20684 Min. :161.2
## 1st Qu.:1.432 1st Qu.:0.7159 1st Qu.: 36714 1st Qu.:215.1
## Median :1.551 Median :0.7644 Median : 45178 Median :238.4
## Mean :1.583 Mean :0.7509 Mean : 53768 Mean :253.1
## 3rd Qu.:1.707 3rd Qu.:0.8105 3rd Qu.: 62294 3rd Qu.:279.4
## Max. :2.430 Max. :0.9114 Max. :263261 Max. :569.4
## Extent Solidity roundness Compactness
## Min. :0.5553 Min. :0.9192 Min. :0.4896 Min. :0.6406
## 1st Qu.:0.7186 1st Qu.:0.9857 1st Qu.:0.8321 1st Qu.:0.7625
## Median :0.7599 Median :0.9883 Median :0.8832 Median :0.8013
## Mean :0.7497 Mean :0.9871 Mean :0.8733 Mean :0.7999
## 3rd Qu.:0.7869 3rd Qu.:0.9900 3rd Qu.:0.9169 3rd Qu.:0.8343
## Max. :0.8662 Max. :0.9947 Max. :0.9907 Max. :0.9873
## ShapeFactor1 ShapeFactor2 ShapeFactor3 ShapeFactor4
## Min. :0.002778 Min. :0.0005642 Min. :0.4103 Min. :0.9477
## 1st Qu.:0.005900 1st Qu.:0.0011535 1st Qu.:0.5814 1st Qu.:0.9937
## Median :0.006645 Median :0.0016935 Median :0.6420 Median :0.9964
## Mean :0.006564 Mean :0.0017159 Mean :0.6436 Mean :0.9951
## 3rd Qu.:0.007271 3rd Qu.:0.0021703 3rd Qu.:0.6960 3rd Qu.:0.9979
## Max. :0.010451 Max. :0.0036650 Max. :0.9748 Max. :0.9997
## ClassBARBUNYA ClassBOMBAY ClassCALI ClassDERMASON
## Min. :0.00000 Min. :0.00000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.0000 1st Qu.:0.0000
## Median :0.00000 Median :0.00000 Median :0.0000 Median :0.0000
## Mean :0.09713 Mean :0.03835 Mean :0.1198 Mean :0.2605
## 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:0.0000 3rd Qu.:1.0000
## Max. :1.00000 Max. :1.00000 Max. :1.0000 Max. :1.0000
## ClassHOROZ ClassSEKER ClassSIRA
## Min. :0.0000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.0000
## Median :0.0000 Median :0.0000 Median :0.0000
## Mean :0.1417 Mean :0.1489 Mean :0.1937
## 3rd Qu.:0.0000 3rd Qu.:0.0000 3rd Qu.:0.0000
## Max. :1.0000 Max. :1.0000 Max. :1.0000
dry_bean_dataset_one_hot_1000_df <- dry_bean_dataset_one_hot_df[sample(nrow(dry_bean_dataset_one_hot_df), size = 1000, replace = FALSE), ]
head(str(dry_bean_dataset_one_hot_1000_df))
## 'data.frame': 1000 obs. of 23 variables:
## $ Area : num 95754 43864 22144 27940 53196 ...
## $ Perimeter : num 1182 799 558 615 905 ...
## $ MajorAxisLength: num 453 303 199 227 364 ...
## $ MinorAxisLength: num 273 184 143 157 187 ...
## $ AspectRation : num 1.66 1.65 1.39 1.45 1.95 ...
## $ Eccentricity : num 0.799 0.794 0.695 0.723 0.859 ...
## $ ConvexArea : num 97441 44336 22445 28256 53781 ...
## $ EquivDiameter : num 349 236 168 189 260 ...
## $ Extent : num 0.749 0.733 0.72 0.808 0.775 ...
## $ Solidity : num 0.983 0.989 0.987 0.989 0.989 ...
## $ roundness : num 0.861 0.863 0.895 0.929 0.817 ...
## $ Compactness : num 0.771 0.779 0.843 0.83 0.715 ...
## $ ShapeFactor1 : num 0.00473 0.00692 0.00899 0.00813 0.00685 ...
## $ ShapeFactor2 : num 0.00103 0.00157 0.0028 0.00238 0.0011 ...
## $ ShapeFactor3 : num 0.595 0.607 0.711 0.689 0.511 ...
## $ ShapeFactor4 : num 0.988 0.998 0.989 0.998 0.996 ...
## $ ClassBARBUNYA : num 0 0 0 0 0 0 0 0 0 0 ...
## $ ClassBOMBAY : num 0 0 0 0 0 0 0 0 0 0 ...
## $ ClassCALI : num 1 0 0 0 0 1 0 0 0 1 ...
## $ ClassDERMASON : num 0 0 1 1 0 0 0 0 1 0 ...
## $ ClassHOROZ : num 0 0 0 0 1 0 0 0 0 0 ...
## $ ClassSEKER : num 0 0 0 0 0 0 0 1 0 0 ...
## $ ClassSIRA : num 0 1 0 0 0 0 1 0 0 0 ...
## NULL
##Persistent Homology of DryBean dataset
# calculate persistent homology for DryBean Dataset
phom_drybean_df <- calculate_homology(dry_bean_dataset_one_hot_1000_df)
# plot barcode for DryBean Dataset
plot_barcode(phom_drybean_df)
# plot persistent diagram of DryBean Dataset
plot_persist(phom_drybean_df)
#####———————————————MAPPER ALGORITHM————————————————
#Prepare Dry Bean dataset for Mapper 1D algorithm
##Two Filter Functions PCA & KDE
#Prepare linear PCA as a filter function by centering and scaling dataset first on all one hot df dataset
b<- prcomp(dry_bean_dataset_one_hot_df, center=TRUE, scale=TRUE)
ts_dry_bean_pca_b <- as.data.frame(predict(b, dry_bean_dataset_one_hot_df))
#Conduct kernel density estimator as a filter function on 4 of 6
filter.kde <- kde(dry_bean_dataset_one_hot_df[,1:4],H=diag(1,nrow = 4),eval.points = dry_bean_dataset_one_hot_df[,1:4])$estimate
###*** dry_bean_dataset PCA Mapper 5 intervals, 60% overlap, 5 bins
##*** dry_bean_dataset PCA Mapper 5 intervals, 60% overlap, 5 bins
m_dry_bean_dataset_5.60.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(ts_dry_bean_pca_b$PC1),
num_intervals = 5,
percent_overlap = 60,
num_bins_when_clustering = 5)
g_dry_bean_dataset_5.60.5 <- graph.adjacency(m_dry_bean_dataset_5.60.5$adjacency, mode="undirected")
## Warning: `graph.adjacency()` was deprecated in igraph 2.0.0.
## ℹ Please use `graph_from_adjacency_matrix()` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
plot(g_dry_bean_dataset_5.60.5, layout = layout.auto(g_dry_bean_dataset_5.60.5))
## Warning: `layout.auto()` was deprecated in igraph 2.0.0.
## ℹ Please use `layout_nicely()` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
head(str(m_dry_bean_dataset_5.60.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_dry_bean_dataset_5.60.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_dry_bean_dataset_5.60.5$points_in_vertex))
## List of 5
## $ : int [1:8547] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:10852] 25 43 108 198 211 272 279 294 369 374 ...
## $ : int [1:5983] 272 1924 1973 1984 2010 2017 2018 2025 2028 2031 ...
## $ : int [1:3037] 272 2216 2236 2283 2346 2383 2407 2421 2432 2437 ...
## $ : int [1:505] 3357 3361 3364 3365 3367 3368 3369 3370 3372 3374 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_dry_bean_dataset_5.60.5$level_of_vertex, na.rm=TRUE)
my_vector = m_dry_bean_dataset_5.60.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_dry_bean_dataset_5.60.5 <- graph.adjacency(m_dry_bean_dataset_5.60.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_dry_bean_dataset_5.60.5$points_in_vertex,
function(x) length(x)))
plot(g_dry_bean_dataset_5.60.5, layout = layout.auto(g_dry_bean_dataset_5.60.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
m_dry_bean_dataset_5.60.5.n1<-m_dry_bean_dataset_5.60.5$points_in_vertex[1]
m_dry_bean_dataset_5.60.5.n1.vec<-as.vector(unlist(m_dry_bean_dataset_5.60.5.n1))
m_dry_bean_dataset_5.60.5.n2<-m_dry_bean_dataset_5.60.5$points_in_vertex[2]
m_dry_bean_dataset_5.60.5.n2.vec<-as.vector(unlist(m_dry_bean_dataset_5.60.5.n2))
m_dry_bean_dataset_5.60.5.n3<-m_dry_bean_dataset_5.60.5$points_in_vertex[3]
m_dry_bean_dataset_5.60.5.n3.vec<-as.vector(unlist(m_dry_bean_dataset_5.60.5.n3))
m_dry_bean_dataset_5.60.5.n4<-m_dry_bean_dataset_5.60.5$points_in_vertex[4]
m_dry_bean_dataset_5.60.5.n4.vec<-as.vector(unlist(m_dry_bean_dataset_5.60.5.n4))
m_dry_bean_dataset_5.60.5.n5<-m_dry_bean_dataset_5.60.5$points_in_vertex[5]
m_dry_bean_dataset_5.60.5.n5.vec<-as.vector(unlist(m_dry_bean_dataset_5.60.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF1 dataset
tda.m_dry_bean_dataset_5.60.5.n1.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.60.5.n1.vec,]
tda.m_dry_bean_dataset_5.60.5.n2.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.60.5.n2.vec,]
tda.m_dry_bean_dataset_5.60.5.n3.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.60.5.n3.vec,]
tda.m_dry_bean_dataset_5.60.5.n4.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.60.5.n4.vec,]
tda.m_dry_bean_dataset_5.60.5.n5.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.60.5.n5.vec,]
###*** dry_bean_dataset Mapper 5 intervals, 50% overlap, 5 bins
m_dry_bean_dataset_5.50.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(ts_dry_bean_pca_b$PC1),
num_intervals = 5,
percent_overlap = 50,
num_bins_when_clustering = 5)
g_dry_bean_dataset_5.50.5 <- graph.adjacency(m_dry_bean_dataset_5.50.5$adjacency, mode="undirected")
plot(g_dry_bean_dataset_5.50.5, layout = layout.auto(g_dry_bean_dataset_5.50.5))
head(str(m_dry_bean_dataset_5.50.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_dry_bean_dataset_5.50.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_dry_bean_dataset_5.50.5$points_in_vertex))
## List of 5
## $ : int [1:7839] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:9515] 272 279 294 402 413 431 433 446 457 548 ...
## $ : int [1:5355] 272 2010 2028 2031 2035 2037 2040 2043 2046 2052 ...
## $ : int [1:1590] 2236 2455 2463 2512 2515 2555 2647 2663 2664 2721 ...
## $ : int [1:417] 3369 3375 3398 3399 3406 3410 3411 3412 3413 3414 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_dry_bean_dataset_5.50.5$level_of_vertex, na.rm=TRUE)
my_vector = m_dry_bean_dataset_5.50.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_dry_bean_dataset_5.50.5 <- graph.adjacency(m_dry_bean_dataset_5.50.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_dry_bean_dataset_5.50.5$points_in_vertex,
function(x) length(x)))
plot(g_dry_bean_dataset_5.50.5, layout = layout.auto(g_dry_bean_dataset_5.50.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
##Extract the ID observations of each mapper output vertex
m_dry_bean_dataset_5.50.5.n1<-m_dry_bean_dataset_5.50.5$points_in_vertex[1]
m_dry_bean_dataset_5.50.5.n1.vec<-as.vector(unlist(m_dry_bean_dataset_5.50.5.n1))
m_dry_bean_dataset_5.50.5.n2<-m_dry_bean_dataset_5.50.5$points_in_vertex[2]
m_dry_bean_dataset_5.50.5.n2.vec<-as.vector(unlist(m_dry_bean_dataset_5.50.5.n2))
m_dry_bean_dataset_5.50.5.n3<-m_dry_bean_dataset_5.50.5$points_in_vertex[3]
m_dry_bean_dataset_5.50.5.n3.vec<-as.vector(unlist(m_dry_bean_dataset_5.50.5.n3))
m_dry_bean_dataset_5.50.5.n4<-m_dry_bean_dataset_5.50.5$points_in_vertex[4]
m_dry_bean_dataset_5.50.5.n4.vec<-as.vector(unlist(m_dry_bean_dataset_5.50.5.n4))
m_dry_bean_dataset_5.50.5.n5<-m_dry_bean_dataset_5.50.5$points_in_vertex[5]
m_dry_bean_dataset_5.50.5.n5.vec<-as.vector(unlist(m_dry_bean_dataset_5.50.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF1 dataset
tda.m_dry_bean_dataset_5.50.5.n1.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.50.5.n1.vec,]
tda.m_dry_bean_dataset_5.50.5.n2.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.50.5.n2.vec,]
tda.m_dry_bean_dataset_5.50.5.n3.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.50.5.n3.vec,]
tda.m_dry_bean_dataset_5.50.5.n4.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.50.5.n4.vec,]
tda.m_dry_bean_dataset_5.50.5.n5.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.50.5.n5.vec,]
##*** dry_bean_dataset Mapper 5 intervals, 40% overlap, 5 bins
m_dry_bean_dataset_5.40.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(ts_dry_bean_pca_b$PC1),
num_intervals = 5,
percent_overlap = 40,
num_bins_when_clustering = 5)
g_dry_bean_dataset_5.40.5 <- graph.adjacency(m_dry_bean_dataset_5.40.5$adjacency, mode="undirected")
plot(g_dry_bean_dataset_5.40.5, layout = layout.auto(g_dry_bean_dataset_5.40.5))
head(str(m_dry_bean_dataset_5.40.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_dry_bean_dataset_5.40.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_dry_bean_dataset_5.40.5$points_in_vertex))
## List of 5
## $ : int [1:6835] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:8024] 272 279 431 433 457 646 667 713 759 798 ...
## $ : int [1:5008] 272 2028 2046 2054 2055 2056 2059 2060 2063 2064 ...
## $ : int [1:894] 2647 2935 2951 2987 3064 3066 3081 3082 3084 3093 ...
## $ : int [1:342] 3375 3424 3427 3428 3430 3435 3437 3450 3453 3456 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_dry_bean_dataset_5.40.5$level_of_vertex, na.rm=TRUE)
my_vector = m_dry_bean_dataset_5.40.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_dry_bean_dataset_5.40.5 <- graph.adjacency(m_dry_bean_dataset_5.40.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_dry_bean_dataset_5.40.5$points_in_vertex,
function(x) length(x)))
plot(g_dry_bean_dataset_5.50.5, layout = layout.auto(g_dry_bean_dataset_5.40.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
m_dry_bean_dataset_5.40.5.n1<-m_dry_bean_dataset_5.40.5$points_in_vertex[1]
m_dry_bean_dataset_5.40.5.n1.vec<-as.vector(unlist(m_dry_bean_dataset_5.40.5.n1))
m_dry_bean_dataset_5.40.5.n2<-m_dry_bean_dataset_5.40.5$points_in_vertex[2]
m_dry_bean_dataset_5.40.5.n2.vec<-as.vector(unlist(m_dry_bean_dataset_5.40.5.n2))
m_dry_bean_dataset_5.40.5.n3<-m_dry_bean_dataset_5.40.5$points_in_vertex[3]
m_dry_bean_dataset_5.40.5.n3.vec<-as.vector(unlist(m_dry_bean_dataset_5.40.5.n3))
m_dry_bean_dataset_5.40.5.n4<-m_dry_bean_dataset_5.40.5$points_in_vertex[4]
m_dry_bean_dataset_5.40.5.n4.vec<-as.vector(unlist(m_dry_bean_dataset_5.40.5.n4))
m_dry_bean_dataset_5.40.5.n5<-m_dry_bean_dataset_5.40.5$points_in_vertex[5]
m_dry_bean_dataset_5.40.5.n5.vec<-as.vector(unlist(m_dry_bean_dataset_5.40.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF1 dataset
tda.m_dry_bean_dataset_5.40.5.n1.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.40.5.n1.vec,]
tda.m_dry_bean_dataset_5.40.5.n2.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.40.5.n2.vec,]
tda.m_dry_bean_dataset_5.40.5.n3.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.40.5.n3.vec,]
tda.m_dry_bean_dataset_5.40.5.n4.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.40.5.n4.vec,]
tda.m_dry_bean_dataset_5.40.5.n5.vec<-Dry_Bean_Dataset[m_dry_bean_dataset_5.40.5.n5.vec,]
##*** dry_bean_dataset Mapper KDE Filter 5 intervals, 60% overlap, 5 bins
m_kde_dry_bean_dataset_5.60.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(filter.kde),
num_intervals = 5,
percent_overlap = 60,
num_bins_when_clustering = 5)
g_kde_dry_bean_dataset_5.60.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.60.5$adjacency, mode="undirected")
plot(g_kde_dry_bean_dataset_5.60.5, layout = layout.auto(g_kde_dry_bean_dataset_5.60.5))
head(str(m_kde_dry_bean_dataset_5.60.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_kde_dry_bean_dataset_5.60.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_kde_dry_bean_dataset_5.60.5$points_in_vertex))
## List of 5
## $ : int [1:9688] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:8239] 1 3 4 6 7 8 9 10 11 12 ...
## $ : int [1:4917] 25 39 43 68 96 102 104 108 114 142 ...
## $ : int [1:2448] 198 211 279 294 307 347 355 360 369 371 ...
## $ : int [1:1349] 402 431 433 438 443 456 457 488 531 536 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_kde_dry_bean_dataset_5.60.5$level_of_vertex, na.rm=TRUE)
my_vector = m_kde_dry_bean_dataset_5.60.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_kde_dry_bean_dataset_5.50.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.60.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_kde_dry_bean_dataset_5.60.5$points_in_vertex,
function(x) length(x)))
plot(g_kde_dry_bean_dataset_5.60.5, layout = layout.auto(g_kde_dry_bean_dataset_5.60.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
##Extract the ID observations of each mapper output vertex
m_kde_dry_bean_dataset_5.60.5.n1<-m_kde_dry_bean_dataset_5.60.5$points_in_vertex[1]
m_kde_dry_bean_dataset_5.60.5.n1.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.60.5.n1))
m_kde_dry_bean_dataset_5.60.5.n2<-m_kde_dry_bean_dataset_5.60.5$points_in_vertex[2]
m_kde_dry_bean_dataset_5.60.5.n2.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.60.5.n2))
m_kde_dry_bean_dataset_5.60.5.n3<-m_kde_dry_bean_dataset_5.60.5$points_in_vertex[3]
m_kde_dry_bean_dataset_5.60.5.n3.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.60.5.n3))
m_kde_dry_bean_dataset_5.60.5.n4<-m_kde_dry_bean_dataset_5.60.5$points_in_vertex[4]
m_kde_dry_bean_dataset_5.60.5.n4.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.60.5.n4))
m_kde_dry_bean_dataset_5.60.5.n5<-m_kde_dry_bean_dataset_5.60.5 $points_in_vertex[5]
m_kde_dry_bean_dataset_5.60.5.n5.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.60.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF1 dataset
tda.m_kde_dry_bean_dataset_5.60.5.n1.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.60.5.n1.vec,]
tda.m_kde_dry_bean_dataset_5.60.5.n2.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.60.5.n2.vec,]
tda.m_kde_dry_bean_dataset_5.60.5.n3.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.60.5.n3.vec,]
tda.m_kde_dry_bean_dataset_5.60.5.n4.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.60.5.n4.vec,]
tda.m_kde_dry_bean_dataset_5.60.5.n5.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.60.5.n5.vec,]
##*** dry_bean_dataset Mapper KDE Filter 5 intervals, 50% overlap, 5 bins
m_kde_dry_bean_dataset_5.50.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(filter.kde),
num_intervals = 5,
percent_overlap = 50,
num_bins_when_clustering = 5)
g_kde_dry_bean_dataset_5.50.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.50.5$adjacency, mode="undirected")
plot(g_kde_dry_bean_dataset_5.50.5, layout = layout.auto(g_kde_dry_bean_dataset_5.50.5))
head(str(m_kde_dry_bean_dataset_5.50.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_kde_dry_bean_dataset_5.50.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_kde_dry_bean_dataset_5.50.5$points_in_vertex))
## List of 5
## $ : int [1:8473] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:7582] 1 3 4 6 7 8 9 10 11 13 ...
## $ : int [1:4149] 25 43 68 96 102 108 151 158 159 162 ...
## $ : int [1:2024] 279 294 369 374 376 385 388 401 402 409 ...
## $ : int [1:989] 431 457 488 536 548 582 587 590 593 605 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_kde_dry_bean_dataset_5.50.5$level_of_vertex, na.rm=TRUE)
my_vector = m_kde_dry_bean_dataset_5.50.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_kde_dry_bean_dataset_5.50.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.50.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_kde_dry_bean_dataset_5.50.5$points_in_vertex,
function(x) length(x)))
plot(g_kde_dry_bean_dataset_5.50.5, layout = layout.auto(g_kde_dry_bean_dataset_5.50.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
##Extract the ID observations of each mapper output vertex
m_kde_dry_bean_dataset_5.50.5.n1<-m_kde_dry_bean_dataset_5.50.5$points_in_vertex[1]
m_kde_dry_bean_dataset_5.50.5.n1.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.50.5.n1))
m_kde_dry_bean_dataset_5.50.5.n2<-m_kde_dry_bean_dataset_5.50.5$points_in_vertex[2]
m_kde_dry_bean_dataset_5.50.5.n2.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.50.5.n2))
m_kde_dry_bean_dataset_5.50.5.n3<-m_kde_dry_bean_dataset_5.50.5$points_in_vertex[3]
m_kde_dry_bean_dataset_5.50.5.n3.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.50.5.n3))
m_kde_dry_bean_dataset_5.50.5.n4<-m_kde_dry_bean_dataset_5.50.5$points_in_vertex[4]
m_kde_dry_bean_dataset_5.50.5.n4.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.50.5.n4))
m_kde_dry_bean_dataset_5.50.5.n5<-m_kde_dry_bean_dataset_5.50.5 $points_in_vertex[5]
m_kde_dry_bean_dataset_5.50.5.n5.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.50.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF1 dataset
tda.m_kde_dry_bean_dataset_5.50.5.n1.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.50.5.n1.vec,]
tda.m_kde_dry_bean_dataset_5.50.5.n2.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.50.5.n2.vec,]
tda.m_kde_dry_bean_dataset_5.50.5.n3.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.50.5.n3.vec,]
tda.m_kde_dry_bean_dataset_5.50.5.n4.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.50.5.n4.vec,]
tda.m_kde_dry_bean_dataset_5.50.5.n5.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.50.5.n5.vec,]
##*** dry_bean_dataset Mapper KDE 5 intervals, 40% overlap, 5 bins
m_kde_dry_bean_dataset_5.40.5 <- mapper1D(
distance_matrix = dist(dry_bean_dataset_one_hot_df),
filter_values = c(filter.kde),
num_intervals = 5,
percent_overlap = 40,
num_bins_when_clustering = 5)
g_kde_dry_bean_dataset_5.40.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.40.5$adjacency, mode="undirected")
plot(g_kde_dry_bean_dataset_5.40.5, layout = layout.auto(g_kde_dry_bean_dataset_5.40.5))
head(str(m_kde_dry_bean_dataset_5.40.5$level_of_vertex))
## int [1:5] 1 2 3 4 5
## NULL
head(str(m_kde_dry_bean_dataset_5.40.5$vertices_in_level))
## List of 5
## $ : num 1
## $ : num 2
## $ : num 3
## $ : num 4
## $ : num 5
## NULL
head(str(m_kde_dry_bean_dataset_5.40.5$points_in_vertex))
## List of 5
## $ : int [1:7503] 1 2 3 4 5 6 7 8 9 10 ...
## $ : int [1:7002] 1 3 4 6 8 9 10 11 13 14 ...
## $ : int [1:3511] 25 108 159 183 197 198 202 206 209 211 ...
## $ : int [1:1759] 294 369 374 376 401 402 409 413 431 433 ...
## $ : int [1:774] 548 593 615 616 618 631 633 638 640 646 ...
## NULL
my_resolution = 100
my_palette = colorRampPalette(c('red','green','lightblue'))
my_max = max(m_kde_dry_bean_dataset_5.40.5$level_of_vertex, na.rm=TRUE)
my_vector = m_kde_dry_bean_dataset_5.40.5$level_of_vertex / my_max
my_colors = my_palette(my_resolution)[as.numeric(cut(
my_vector, breaks=my_resolution))]
g_kde_dry_bean_dataset_5.40.5 <- graph.adjacency(m_kde_dry_bean_dataset_5.40.5$adjacency, mode="undirected")
vertex_size <- unlist(lapply(m_kde_dry_bean_dataset_5.40.5$points_in_vertex,
function(x) length(x)))
plot(g_kde_dry_bean_dataset_5.40.5, layout = layout.auto(g_kde_dry_bean_dataset_5.40.5),
vertex.size = 30*log(vertex_size)/
max(log(vertex_size)),
vertex.color = my_colors)
##Extract the ID observations of each mapper output vertex
m_kde_dry_bean_dataset_5.40.5.n1<-m_kde_dry_bean_dataset_5.40.5$points_in_vertex[1]
m_kde_dry_bean_dataset_5.40.5.n1.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.40.5.n1))
m_kde_dry_bean_dataset_5.40.5.n2<-m_kde_dry_bean_dataset_5.40.5$points_in_vertex[2]
m_kde_dry_bean_dataset_5.40.5.n2.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.40.5.n2))
m_kde_dry_bean_dataset_5.40.5.n3<-m_kde_dry_bean_dataset_5.40.5$points_in_vertex[3]
m_kde_dry_bean_dataset_5.40.5.n3.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.40.5.n3))
m_kde_dry_bean_dataset_5.40.5.n4<-m_kde_dry_bean_dataset_5.40.5$points_in_vertex[4]
m_kde_dry_bean_dataset_5.40.5.n4.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.40.5.n4))
m_kde_dry_bean_dataset_5.40.5.n5<-m_kde_dry_bean_dataset_5.40.5 $points_in_vertex[5]
m_kde_dry_bean_dataset_5.40.5.n5.vec<-as.vector(unlist(m_kde_dry_bean_dataset_5.40.5.n5))
##map the ID’s of each Mapper vertex point to the actual dry_bean_dataset One Hot DF4 dataset
tda.m_kde_dry_bean_dataset_5.40.5.n1.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.40.5.n1.vec,]
tda.m_kde_dry_bean_dataset_5.40.5.n2.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.40.5.n2.vec,]
tda.m_kde_dry_bean_dataset_5.40.5.n3.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.40.5.n3.vec,]
tda.m_kde_dry_bean_dataset_5.40.5.n4.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.40.5.n4.vec,]
tda.m_kde_dry_bean_dataset_5.40.5.n5.vec<-Dry_Bean_Dataset[m_kde_dry_bean_dataset_5.40.5.n5.vec,]
library(caret)
trainIndex <- createDataPartition(Dry_Bean_Dataset$Class, p = .7,
list = FALSE,
times = 1)
head(trainIndex)
## Resample1
## [1,] 2
## [2,] 4
## [3,] 5
## [4,] 6
## [5,] 7
## [6,] 8
Dry_Bean_DatasetTrain <- Dry_Bean_Dataset[ trainIndex,]
Dry_Bean_DatasetTest <- Dry_Bean_Dataset[-trainIndex,]
#Train Control: k-Fold Cross-validation basis for all models
fitControl <- trainControl(## 10-fold CV
method = "cv",
number = 3)
#Non-TDA-Assited
rfGrid<-expand.grid(mtry = (1:20)*50)
#Random Forest
dryBeanRfFit <- train(as.factor(Class) ~ ., data = Dry_Bean_DatasetTrain,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
dryBeanRfFit
## Random Forest
##
## 9531 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6354, 6355, 6353
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9265562 0.9111474
## 100 0.9256121 0.9100126
## 150 0.9251923 0.9094962
## 200 0.9259265 0.9103820
## 250 0.9262413 0.9107681
## 300 0.9252971 0.9096266
## 350 0.9251921 0.9095050
## 400 0.9256119 0.9100022
## 450 0.9247724 0.9089949
## 500 0.9261365 0.9106455
## 550 0.9260314 0.9105091
## 600 0.9246677 0.9088703
## 650 0.9252970 0.9096301
## 700 0.9250874 0.9093706
## 750 0.9250873 0.9093743
## 800 0.9258219 0.9102635
## 850 0.9252970 0.9096184
## 900 0.9257170 0.9101351
## 950 0.9256117 0.9100080
## 1000 0.9243527 0.9084797
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 50.
dryBeanRfFit$resample
## Accuracy Kappa Resample
## 1 0.9250866 0.9093960 Fold1
## 2 0.9238515 0.9078857 Fold3
## 3 0.9307305 0.9161604 Fold2
db_rf_fit_re<-dryBeanRfFit$resample[1]
summary(dryBeanRfFit)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 9531 factor numeric
## err.rate 4000 -none- numeric
## confusion 56 -none- numeric
## votes 66717 matrix numeric
## oob.times 9531 -none- numeric
## classes 7 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 9531 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 7 -none- character
## param 1 -none- list
vip(dryBeanRfFit,25) + ggtitle("non-TDA-Assisted: RF")
# Predict outcome using model from training data based on testing data
predictions <- predict(dryBeanRfFit, newdata = Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_rf_cf<-confusionMatrix(data=predictions, as.factor(Dry_Bean_DatasetTest$Class))
db_rf_cf
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 363 0 15 0 3 4 7
## BOMBAY 1 156 0 0 0 0 0
## CALI 19 0 461 0 12 0 1
## DERMASON 1 0 0 969 6 17 90
## HOROZ 5 0 7 2 548 0 12
## SEKER 1 0 1 21 0 573 5
## SIRA 6 0 5 71 9 14 675
##
## Overall Statistics
##
## Accuracy : 0.9179
## 95% CI : (0.909, 0.9261)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9007
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91667 1.00000 0.9427 0.9116
## Specificity 0.99213 0.99975 0.9911 0.9622
## Pos Pred Value 0.92602 0.99363 0.9351 0.8947
## Neg Pred Value 0.99105 1.00000 0.9922 0.9686
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08897 0.03824 0.1130 0.2375
## Detection Prevalence 0.09608 0.03848 0.1208 0.2654
## Balanced Accuracy 0.95440 0.99987 0.9669 0.9369
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9481 0.9424 0.8544
## Specificity 0.9926 0.9919 0.9681
## Pos Pred Value 0.9547 0.9534 0.8654
## Neg Pred Value 0.9914 0.9899 0.9652
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1343 0.1404 0.1654
## Detection Prevalence 0.1407 0.1473 0.1912
## Balanced Accuracy 0.9703 0.9672 0.9113
db_rf_cf$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9178922 0.9006758 0.9090433 0.9261370 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_rf_cf_ov_acc<-db_rf_cf$overall[1]
db_rf_cf$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9166667 0.9921281 0.9260204 0.9910521 0.9260204
## Class: BOMBAY 1.0000000 0.9997452 0.9936306 1.0000000 0.9936306
## Class: CALI 0.9427403 0.9910888 0.9350913 0.9921940 0.9350913
## Class: DERMASON 0.9115710 0.9622141 0.8947368 0.9686353 0.8947368
## Class: HOROZ 0.9480969 0.9925757 0.9547038 0.9914432 0.9547038
## Class: SEKER 0.9424342 0.9919355 0.9534110 0.9899396 0.9534110
## Class: SIRA 0.8544304 0.9680851 0.8653846 0.9651515 0.8653846
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9166667 0.9213198 0.09705882 0.08897059
## Class: BOMBAY 1.0000000 0.9968051 0.03823529 0.03823529
## Class: CALI 0.9427403 0.9389002 0.11985294 0.11299020
## Class: DERMASON 0.9115710 0.9030755 0.26053922 0.23750000
## Class: HOROZ 0.9480969 0.9513889 0.14166667 0.13431373
## Class: SEKER 0.9424342 0.9478908 0.14901961 0.14044118
## Class: SIRA 0.8544304 0.8598726 0.19362745 0.16544118
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09607843 0.9543974
## Class: BOMBAY 0.03848039 0.9998726
## Class: CALI 0.12083333 0.9669146
## Class: DERMASON 0.26544118 0.9368926
## Class: HOROZ 0.14068627 0.9703363
## Class: SEKER 0.14730392 0.9671848
## Class: SIRA 0.19117647 0.9112577
db_rf_cf_pre_rec_f1<-db_rf_cf$byClass[5:7]
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_PC_5.50.5_n1_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_PC_5.50.5_n1_RfFit0
## Random Forest
##
## 7839 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5224, 5226, 5228
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9109577 0.8623916
## 100 0.9103201 0.8614161
## 150 0.9103200 0.8614826
## 200 0.9104481 0.8615998
## 250 0.9104481 0.8616920
## 300 0.9118504 0.8638404
## 350 0.9107030 0.8620755
## 400 0.9096821 0.8605097
## 450 0.9105749 0.8618106
## 500 0.9101924 0.8612533
## 550 0.9109578 0.8624083
## 600 0.9115956 0.8634139
## 650 0.9107035 0.8620544
## 700 0.9105756 0.8618413
## 750 0.9115960 0.8634190
## 800 0.9103197 0.8613922
## 850 0.9101927 0.8612581
## 900 0.9105752 0.8618085
## 950 0.9096826 0.8604303
## 1000 0.9107029 0.8620331
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 300.
DryBean_TDA_PC_5.50.5_n1_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9116635 0.8636336 Fold1
## 2 0.9092302 0.8595162 Fold3
## 3 0.9146575 0.8683713 Fold2
db_tda_pc_5.50.5_n1_rf_fit0_re<-DryBean_TDA_PC_5.50.5_n1_RfFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 7839 factor numeric
## err.rate 3500 -none- numeric
## confusion 42 -none- numeric
## votes 47034 matrix numeric
## oob.times 7839 -none- numeric
## classes 6 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 7839 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 6 -none- character
## param 1 -none- list
vip(DryBean_TDA_PC_5.50.5_n1_RfFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n1_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_PC_5.50.5_n1_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n1_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n1_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n1_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 253 22 121 0 70 0 1
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 1 0 0 0 0
## DERMASON 0 0 0 1063 198 0 1
## HOROZ 0 0 0 0 8 0 0
## SEKER 138 134 362 0 11 608 18
## SIRA 5 0 5 0 291 0 770
##
## Overall Statistics
##
## Accuracy : 0.6625
## 95% CI : (0.6478, 0.677)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5837
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.63889 0.00000 0.0020450 1.0000
## Specificity 0.94191 1.00000 1.0000000 0.9340
## Pos Pred Value 0.54176 NaN 1.0000000 0.8423
## Neg Pred Value 0.96042 0.96176 0.8803628 1.0000
## Prevalence 0.09706 0.03824 0.1198529 0.2605
## Detection Rate 0.06201 0.00000 0.0002451 0.2605
## Detection Prevalence 0.11446 0.00000 0.0002451 0.3093
## Balanced Accuracy 0.79040 0.50000 0.5010225 0.9670
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.013841 1.0000 0.9747
## Specificity 1.000000 0.8090 0.9085
## Pos Pred Value 1.000000 0.4784 0.7190
## Neg Pred Value 0.860020 1.0000 0.9934
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.001961 0.1490 0.1887
## Detection Prevalence 0.001961 0.3115 0.2625
## Balanced Accuracy 0.506920 0.9045 0.9416
db_tda_pc_5.50.5_n1_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 253 22 121 0 70 0 1
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 1 0 0 0 0
## DERMASON 0 0 0 1063 198 0 1
## HOROZ 0 0 0 0 8 0 0
## SEKER 138 134 362 0 11 608 18
## SIRA 5 0 5 0 291 0 770
##
## Overall Statistics
##
## Accuracy : 0.6625
## 95% CI : (0.6478, 0.677)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5837
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.63889 0.00000 0.0020450 1.0000
## Specificity 0.94191 1.00000 1.0000000 0.9340
## Pos Pred Value 0.54176 NaN 1.0000000 0.8423
## Neg Pred Value 0.96042 0.96176 0.8803628 1.0000
## Prevalence 0.09706 0.03824 0.1198529 0.2605
## Detection Rate 0.06201 0.00000 0.0002451 0.2605
## Detection Prevalence 0.11446 0.00000 0.0002451 0.3093
## Balanced Accuracy 0.79040 0.50000 0.5010225 0.9670
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.013841 1.0000 0.9747
## Specificity 1.000000 0.8090 0.9085
## Pos Pred Value 1.000000 0.4784 0.7190
## Neg Pred Value 0.860020 1.0000 0.9934
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.001961 0.1490 0.1887
## Detection Prevalence 0.001961 0.3115 0.2625
## Balanced Accuracy 0.506920 0.9045 0.9416
db_tda_pc_5.50.5_n1_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6625000 0.5837164 0.6477557 0.6770113 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n1_rf_cf0_ov_acc<-db_tda_pc_5.50.5_n1_rf_cf0$overall[1]
db_tda_pc_5.50.5_n1_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.63888889 0.9419110 0.5417559 0.9604207 0.5417559
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.00204499 1.0000000 1.0000000 0.8803628 1.0000000
## Class: DERMASON 1.00000000 0.9340404 0.8423138 1.0000000 0.8423138
## Class: HOROZ 0.01384083 1.0000000 1.0000000 0.8600196 1.0000000
## Class: SEKER 1.00000000 0.8090438 0.4783635 1.0000000 0.4783635
## Class: SIRA 0.97468354 0.9085106 0.7189542 0.9933533 0.7189542
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.63888889 0.586326767 0.09705882 0.062009804
## Class: BOMBAY 0.00000000 NA 0.03823529 0.000000000
## Class: CALI 0.00204499 0.004081633 0.11985294 0.000245098
## Class: DERMASON 1.00000000 0.914408602 0.26053922 0.260539216
## Class: HOROZ 0.01384083 0.027303754 0.14166667 0.001960784
## Class: SEKER 1.00000000 0.647152741 0.14901961 0.149019608
## Class: SIRA 0.97468354 0.827512090 0.19362745 0.188725490
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.114460784 0.7903999
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000245098 0.5010225
## Class: DERMASON 0.309313725 0.9670202
## Class: HOROZ 0.001960784 0.5069204
## Class: SEKER 0.311519608 0.9045219
## Class: SIRA 0.262500000 0.9415971
db_tda_pc_5.50.5_n1_rf_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n1_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_rf_n1_3_fold<-(db_rf_fit_re-db_tda_pc_5.50.5_n1_rf_fit0_re)
diff_drybean_tda_pca_5.50.5_rf_n1_3_fold
## Accuracy
## 1 0.01342308
## 2 0.01462130
## 3 0.01607300
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n1_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n1_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1694
##
## $winRight
## [1] 0.8306
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n1_3_fold
## $left
## [1] 0.0006398568
##
## $rope
## [1] 0.01614588
##
## $right
## [1] 0.9832143
# Rope Plot
plot(rope(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold),c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold))
#bf_tda_pca_5.50.5_rf.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_rf_n1_3_fold)
## t = 19.195, df = 2, p-value = 0.002703
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.01140940 0.01800219
## sample estimates:
## mean of x
## 0.01470579
### Test set diff
diff_drybean_tda_pca_5.50.5_rf.n1_test<-(db_rf_cf_ov_acc-db_tda_pc_5.50.5_n1_rf_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_rf.n1_test
## Accuracy
## 0.2553922
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n1_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n1_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n1_test$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n1_test$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n1_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1606333
##
## $winRight
## [1] 0.8393667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n1_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_rf.n1_test)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf.n1_test)) #bf_tda_pca_5.50.5_rf.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n1_test))
##Node2
DryBean_TDA_PC_5.50.5_n2_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 50 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 100 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 150 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 200 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 250 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 300 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 350 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 400 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 450 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 500 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 550 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 600 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 650 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 700 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 750 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 800 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 850 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 900 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry= 950 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold3: mtry=1000 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Can't have empty classes in y.
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_PC_5.50.5_n2_RfFit0
## Random Forest
##
## 9515 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6344, 6343, 6343
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9009937 0.8750669
## 100 0.8997325 0.8734818
## 150 0.9020974 0.8764675
## 200 0.9009939 0.8750450
## 250 0.9006784 0.8746598
## 300 0.9013092 0.8754739
## 350 0.9005210 0.8744799
## 400 0.9000480 0.8738692
## 450 0.9008360 0.8748766
## 500 0.9009938 0.8750842
## 550 0.9011517 0.8752628
## 600 0.9014668 0.8756717
## 650 0.9008361 0.8748736
## 700 0.9002058 0.8740720
## 750 0.9009939 0.8750681
## 800 0.9013092 0.8754688
## 850 0.9013091 0.8754314
## 900 0.9003631 0.8742694
## 950 0.9011514 0.8752643
## 1000 0.9020974 0.8764672
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 150.
DryBean_TDA_PC_5.50.5_n2_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.8984868 0.871988 Fold2
## 2 0.9057080 0.880947 Fold1
## 3 NA NA Fold3
db_tda_pc_5.50.5_n2_rf_fit0_re<-DryBean_TDA_PC_5.50.5_n2_RfFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n2_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 9515 factor numeric
## err.rate 4000 -none- numeric
## confusion 56 -none- numeric
## votes 66605 matrix numeric
## oob.times 9515 -none- numeric
## classes 7 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 9515 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 7 -none- character
## param 1 -none- list
vip(DryBean_TDA_PC_5.50.5_n2_RfFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n2_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_PC_5.50.5_n2_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n2_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n2_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n2_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 394 27 2 0 3 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 1 128 472 0 7 0 0
## DERMASON 0 0 0 1059 0 203 0
## HOROZ 1 1 15 0 568 0 0
## SEKER 0 0 0 0 0 376 0
## SIRA 0 0 0 4 0 29 790
##
## Overall Statistics
##
## Accuracy : 0.8968
## 95% CI : (0.8871, 0.906)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8739
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.99495 0.00000 0.9652 0.9962
## Specificity 0.99131 1.00000 0.9621 0.9327
## Pos Pred Value 0.92488 NaN 0.7763 0.8391
## Neg Pred Value 0.99945 0.96176 0.9951 0.9986
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09657 0.00000 0.1157 0.2596
## Detection Prevalence 0.10441 0.00000 0.1490 0.3093
## Balanced Accuracy 0.99313 0.50000 0.9637 0.9645
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9827 0.61842 1.0000
## Specificity 0.9951 1.00000 0.9900
## Pos Pred Value 0.9709 1.00000 0.9599
## Neg Pred Value 0.9971 0.93737 1.0000
## Prevalence 0.1417 0.14902 0.1936
## Detection Rate 0.1392 0.09216 0.1936
## Detection Prevalence 0.1434 0.09216 0.2017
## Balanced Accuracy 0.9889 0.80921 0.9950
db_tda_pc_5.50.5_n2_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 394 27 2 0 3 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 1 128 472 0 7 0 0
## DERMASON 0 0 0 1059 0 203 0
## HOROZ 1 1 15 0 568 0 0
## SEKER 0 0 0 0 0 376 0
## SIRA 0 0 0 4 0 29 790
##
## Overall Statistics
##
## Accuracy : 0.8968
## 95% CI : (0.8871, 0.906)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8739
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.99495 0.00000 0.9652 0.9962
## Specificity 0.99131 1.00000 0.9621 0.9327
## Pos Pred Value 0.92488 NaN 0.7763 0.8391
## Neg Pred Value 0.99945 0.96176 0.9951 0.9986
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09657 0.00000 0.1157 0.2596
## Detection Prevalence 0.10441 0.00000 0.1490 0.3093
## Balanced Accuracy 0.99313 0.50000 0.9637 0.9645
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9827 0.61842 1.0000
## Specificity 0.9951 1.00000 0.9900
## Pos Pred Value 0.9709 1.00000 0.9599
## Neg Pred Value 0.9971 0.93737 1.0000
## Prevalence 0.1417 0.14902 0.1936
## Detection Rate 0.1392 0.09216 0.1936
## Detection Prevalence 0.1434 0.09216 0.2017
## Balanced Accuracy 0.9889 0.80921 0.9950
db_tda_pc_5.50.5_n2_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8968137 0.8739038 0.8870710 0.9059839 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n2_rf_cf0_ov_acc<-db_tda_pc_5.50.5_n2_rf_cf0$overall[1]
db_tda_pc_5.50.5_n2_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9949495 0.9913138 0.9248826 0.9994527 0.9248826
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.9652352 0.9621275 0.7763158 0.9951037 0.7763158
## Class: DERMASON 0.9962371 0.9327146 0.8391442 0.9985806 0.8391442
## Class: HOROZ 0.9826990 0.9951456 0.9709402 0.9971388 0.9709402
## Class: SEKER 0.6184211 1.0000000 1.0000000 0.9373650 1.0000000
## Class: SIRA 1.0000000 0.9899696 0.9599028 1.0000000 0.9599028
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9949495 0.9586375 0.09705882 0.09656863
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.9652352 0.8605287 0.11985294 0.11568627
## Class: DERMASON 0.9962371 0.9109677 0.26053922 0.25955882
## Class: HOROZ 0.9826990 0.9767842 0.14166667 0.13921569
## Class: SEKER 0.6184211 0.7642276 0.14901961 0.09215686
## Class: SIRA 1.0000000 0.9795412 0.19362745 0.19362745
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.10441176 0.9931316
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.14901961 0.9636814
## Class: DERMASON 0.30931373 0.9644758
## Class: HOROZ 0.14338235 0.9889223
## Class: SEKER 0.09215686 0.8092105
## Class: SIRA 0.20171569 0.9949848
db_tda_pc_5.50.5_n2_rf_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n2_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_rf_n2_3_fold<-(db_rf_fit_re-db_tda_pc_5.50.5_n2_rf_fit0_re)
diff_drybean_tda_pca_5.50.5_rf_n2_3_fold
## Accuracy
## 1 0.0265998
## 2 0.0181435
## 3 NA
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold
## $probLeft
## [1] NA
##
## $probRope
## [1] NA
##
## $probRight
## [1] NA
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n2_3_fold_odds.left
## [1] NA
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_rf.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_rf.n2_3_fold
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n2_3_fold
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold),c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold))
#bf_tda_pca_5.50.5_rf.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_rf_n2_3_fold)
## t = 5.2911, df = 1, p-value = 0.1189
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.03135209 0.07609539
## sample estimates:
## mean of x
## 0.02237165
### Test set diff
diff_drybean_tda_pca_5.50.5_rf.n2_test<-(db_rf_cf_ov_acc-db_tda_pc_5.50.5_n2_rf_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_rf.n2_test
## Accuracy
## 0.02107843
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n2_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n2_test$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n2_test$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1619
##
## $winRight
## [1] 0.8381
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test),c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test)) #bf_tda_pca_5.50.5_rf.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test))
##Node3
DryBean_TDA_PC_5.50.5_n3_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_PC_5.50.5_n3_RfFit0
## Random Forest
##
## 5355 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 3569, 3571, 3570
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9363235 0.9111979
## 100 0.9381909 0.9137930
## 150 0.9372577 0.9125095
## 200 0.9370709 0.9122286
## 250 0.9380044 0.9135214
## 300 0.9380042 0.9135431
## 350 0.9361370 0.9109204
## 400 0.9374440 0.9127344
## 450 0.9374440 0.9127340
## 500 0.9380046 0.9134972
## 550 0.9381910 0.9138017
## 600 0.9381909 0.9137758
## 650 0.9381914 0.9137947
## 700 0.9365106 0.9114379
## 750 0.9368839 0.9119667
## 800 0.9389375 0.9148460
## 850 0.9380049 0.9135287
## 900 0.9380047 0.9135499
## 950 0.9378174 0.9132780
## 1000 0.9372577 0.9124974
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 800.
DryBean_TDA_PC_5.50.5_n3_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9467489 0.9258106 Fold2
## 2 0.9361702 0.9108514 Fold1
## 3 0.9338936 0.9078760 Fold3
db_tda_pc_5.50.5_n3_rf_fit0_re<-DryBean_TDA_PC_5.50.5_n3_RfFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n3_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 5355 factor numeric
## err.rate 4000 -none- numeric
## confusion 56 -none- numeric
## votes 37485 matrix numeric
## oob.times 5355 -none- numeric
## classes 7 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 5355 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 7 -none- character
## param 1 -none- list
vip(DryBean_TDA_PC_5.50.5_n3_RfFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n3_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_PC_5.50.5_n3_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n3_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n3_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n3_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 392 1 1 12 0 156 15
## BOMBAY 0 104 0 0 0 0 0
## CALI 0 51 488 0 0 0 1
## DERMASON 0 0 0 1 0 0 0
## HOROZ 0 0 0 829 576 3 49
## SEKER 0 0 0 0 0 0 0
## SIRA 4 0 0 221 2 449 725
##
## Overall Statistics
##
## Accuracy : 0.5603
## 95% CI : (0.5449, 0.5756)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4841
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.98990 0.66667 0.9980 0.0009407
## Specificity 0.94978 1.00000 0.9855 1.0000000
## Pos Pred Value 0.67938 1.00000 0.9037 1.0000000
## Neg Pred Value 0.99886 0.98692 0.9997 0.7396421
## Prevalence 0.09706 0.03824 0.1199 0.2605392
## Detection Rate 0.09608 0.02549 0.1196 0.0002451
## Detection Prevalence 0.14142 0.02549 0.1324 0.0002451
## Balanced Accuracy 0.96984 0.83333 0.9917 0.5004704
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9965 0.000 0.9177
## Specificity 0.7484 1.000 0.7945
## Pos Pred Value 0.3953 NaN 0.5175
## Neg Pred Value 0.9992 0.851 0.9757
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1412 0.000 0.1777
## Detection Prevalence 0.3571 0.000 0.3434
## Balanced Accuracy 0.8725 0.500 0.8561
db_tda_pc_5.50.5_n3_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 392 1 1 12 0 156 15
## BOMBAY 0 104 0 0 0 0 0
## CALI 0 51 488 0 0 0 1
## DERMASON 0 0 0 1 0 0 0
## HOROZ 0 0 0 829 576 3 49
## SEKER 0 0 0 0 0 0 0
## SIRA 4 0 0 221 2 449 725
##
## Overall Statistics
##
## Accuracy : 0.5603
## 95% CI : (0.5449, 0.5756)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4841
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.98990 0.66667 0.9980 0.0009407
## Specificity 0.94978 1.00000 0.9855 1.0000000
## Pos Pred Value 0.67938 1.00000 0.9037 1.0000000
## Neg Pred Value 0.99886 0.98692 0.9997 0.7396421
## Prevalence 0.09706 0.03824 0.1199 0.2605392
## Detection Rate 0.09608 0.02549 0.1196 0.0002451
## Detection Prevalence 0.14142 0.02549 0.1324 0.0002451
## Balanced Accuracy 0.96984 0.83333 0.9917 0.5004704
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9965 0.000 0.9177
## Specificity 0.7484 1.000 0.7945
## Pos Pred Value 0.3953 NaN 0.5175
## Neg Pred Value 0.9992 0.851 0.9757
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1412 0.000 0.1777
## Detection Prevalence 0.3571 0.000 0.3434
## Balanced Accuracy 0.8725 0.500 0.8561
db_tda_pc_5.50.5_n3_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.5602941 0.4840912 0.5449027 0.5755991 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n3_rf_cf0_ov_acc<-db_tda_pc_5.50.5_n3_rf_cf0$overall[1]
db_tda_pc_5.50.5_n3_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0.9898989899 0.9497828 0.6793761 0.9988581
## Class: BOMBAY 0.6666666667 1.0000000 1.0000000 0.9869215
## Class: CALI 0.9979550102 0.9855194 0.9037037 0.9997175
## Class: DERMASON 0.0009407338 1.0000000 1.0000000 0.7396421
## Class: HOROZ 0.9965397924 0.7484295 0.3953329 0.9992375
## Class: SEKER 0.0000000000 1.0000000 NaN 0.8509804
## Class: SIRA 0.9177215190 0.7945289 0.5174875 0.9757372
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.6793761 0.9898989899 0.805755396 0.09705882 0.096078431
## Class: BOMBAY 1.0000000 0.6666666667 0.800000000 0.03823529 0.025490196
## Class: CALI 0.9037037 0.9979550102 0.948493683 0.11985294 0.119607843
## Class: DERMASON 1.0000000 0.0009407338 0.001879699 0.26053922 0.000245098
## Class: HOROZ 0.3953329 0.9965397924 0.566093366 0.14166667 0.141176471
## Class: SEKER NA 0.0000000000 NA 0.14901961 0.000000000
## Class: SIRA 0.5174875 0.9177215190 0.661798266 0.19362745 0.177696078
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.141421569 0.9698409
## Class: BOMBAY 0.025490196 0.8333333
## Class: CALI 0.132352941 0.9917372
## Class: DERMASON 0.000245098 0.5004704
## Class: HOROZ 0.357107843 0.8724846
## Class: SEKER 0.000000000 0.5000000
## Class: SIRA 0.343382353 0.8561252
db_tda_pc_5.50.5_n3_rf_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n3_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_rf_n3_3_fold<-(db_rf_fit_re-db_tda_pc_5.50.5_n3_rf_fit0_re)
diff_drybean_tda_pca_5.50.5_rf_n3_3_fold
## Accuracy
## 1 -0.021662319
## 2 -0.012318734
## 3 -0.003163079
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold
## $probLeft
## [1] 0.5
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n3_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n3_3_fold
## $winLeft
## [1] 0.6055
##
## $winRope
## [1] 0.3945
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n2_3_fold
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
plot(rope(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold),c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold))
#bf_tda_pca_5.50.5_rf.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_rf_n3_3_fold)
## t = -2.3185, df = 2, p-value = 0.1463
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.03535910 0.01059635
## sample estimates:
## mean of x
## -0.01238138
### Test set diff
diff_drybean_tda_pca_5.50.5_rf.n3_test<-(db_rf_cf_ov_acc-db_tda_pc_5.50.5_n3_rf_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_rf.n3_test
## Accuracy
## 0.357598
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n3_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n3_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n3_test$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n3_test$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n3_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1615
##
## $winRight
## [1] 0.8385
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n3_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_rf.n3_test))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf.n3_test)) #bf_tda_pca_5.50.5_rf.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n2_test)
##Node4
DryBean_TDA_PC_5.50.5_n4_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_PC_5.50.5_n4_RfFit0
## Random Forest
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1060, 1060, 1060
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9704403 0.9569691
## 100 0.9691824 0.9551499
## 150 0.9691824 0.9551096
## 200 0.9710692 0.9578962
## 250 0.9698113 0.9560599
## 300 0.9698113 0.9560793
## 350 0.9691824 0.9551499
## 400 0.9698113 0.9560148
## 450 0.9685535 0.9542196
## 500 0.9685535 0.9542536
## 550 0.9691824 0.9551407
## 600 0.9691824 0.9551167
## 650 0.9710692 0.9579371
## 700 0.9691824 0.9551557
## 750 0.9691824 0.9551436
## 800 0.9691824 0.9551379
## 850 0.9710692 0.9578851
## 900 0.9691824 0.9551606
## 950 0.9691824 0.9551173
## 1000 0.9685535 0.9542314
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 200.
DryBean_TDA_PC_5.50.5_n4_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9698113 0.9560312 Fold3
## 2 0.9698113 0.9562056 Fold2
## 3 0.9735849 0.9614517 Fold1
db_tda_pc_5.50.5_n4_rf_fit0_re<-DryBean_TDA_PC_5.50.5_n4_RfFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n4_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 1590 factor numeric
## err.rate 2500 -none- numeric
## confusion 20 -none- numeric
## votes 6360 matrix numeric
## oob.times 1590 -none- numeric
## classes 4 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1590 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 4 -none- character
## param 1 -none- list
vip(DryBean_TDA_PC_5.50.5_n4_RfFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n4_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_PC_5.50.5_n4_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n4_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n4_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n4_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 272 0 4 0 0 4 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 73 0 469 0 13 102 9
## DERMASON 0 0 0 0 0 0 0
## HOROZ 51 0 16 1063 565 502 781
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3583
## 95% CI : (0.3436, 0.3733)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2615
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.68687 1.00000 0.9591 0.0000
## Specificity 0.99783 1.00000 0.9451 1.0000
## Pos Pred Value 0.97143 1.00000 0.7042 NaN
## Neg Pred Value 0.96737 1.00000 0.9941 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.06667 0.03824 0.1150 0.0000
## Detection Prevalence 0.06863 0.03824 0.1632 0.0000
## Balanced Accuracy 0.84235 1.00000 0.9521 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9775 0.000 0.0000
## Specificity 0.3110 1.000 1.0000
## Pos Pred Value 0.1897 NaN NaN
## Neg Pred Value 0.9882 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1385 0.000 0.0000
## Detection Prevalence 0.7299 0.000 0.0000
## Balanced Accuracy 0.6442 0.500 0.5000
db_tda_pc_5.50.5_n4_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 272 0 4 0 0 4 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 73 0 469 0 13 102 9
## DERMASON 0 0 0 0 0 0 0
## HOROZ 51 0 16 1063 565 502 781
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3583
## 95% CI : (0.3436, 0.3733)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2615
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.68687 1.00000 0.9591 0.0000
## Specificity 0.99783 1.00000 0.9451 1.0000
## Pos Pred Value 0.97143 1.00000 0.7042 NaN
## Neg Pred Value 0.96737 1.00000 0.9941 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.06667 0.03824 0.1150 0.0000
## Detection Prevalence 0.06863 0.03824 0.1632 0.0000
## Balanced Accuracy 0.84235 1.00000 0.9521 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9775 0.000 0.0000
## Specificity 0.3110 1.000 1.0000
## Pos Pred Value 0.1897 NaN NaN
## Neg Pred Value 0.9882 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1385 0.000 0.0000
## Detection Prevalence 0.7299 0.000 0.0000
## Balanced Accuracy 0.6442 0.500 0.5000
db_tda_pc_5.50.5_n4_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.583333e-01 2.615270e-01 3.436031e-01 3.732667e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 2.937776e-43 NaN
db_tda_pc_5.50.5_n4_rf_cf0_ov_acc<-db_tda_pc_5.50.5_n4_rf_cf0$overall[1]
db_tda_pc_5.50.5_n4_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.6868687 0.9978284 0.9714286 0.9673684 0.9714286
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9591002 0.9451406 0.7042042 0.9941418 0.7042042
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9775087 0.3109652 0.1897246 0.9882033 0.1897246
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.6868687 0.8047337 0.09705882 0.06666667
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9591002 0.8121212 0.11985294 0.11495098
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9775087 0.3177728 0.14166667 0.13848039
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.06862745 0.8423486
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.16323529 0.9521204
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.72990196 0.6442369
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
db_tda_pc_5.50.5_n4_rf_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n4_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_rf_n4_3_fold<-(db_rf_fit_re-db_tda_pc_5.50.5_n4_rf_fit0_re)
diff_drybean_tda_pca_5.50.5_rf_n4_3_fold
## Accuracy
## 1 -0.04472476
## 2 -0.04595984
## 3 -0.04285443
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n4_3_fold
## $winLeft
## [1] 0.9906
##
## $winRope
## [1] 0.0094
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n4_3_fold
## $left
## [1] 0.9995446
##
## $rope
## [1] 0.0002727294
##
## $right
## [1] 0.0001827024
# Rope Plot
plot(rope(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold),c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold))
#bf_tda_pca_5.50.5_rf.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_rf_n4_3_fold)
## t = -49.312, df = 2, p-value = 0.000411
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.04839696 -0.04062906
## sample estimates:
## mean of x
## -0.04451301
### Test set diff
diff_drybean_tda_pca_5.50.5_rf.n4_test<-(db_rf_cf_ov_acc-db_tda_pc_5.50.5_n4_rf_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_rf.n4_test
## Accuracy
## 0.5595588
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n4_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n4_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n4_test$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n4_test$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n4_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1571333
##
## $winRight
## [1] 0.8428667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n4_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_rf.n4_test))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf.n4_test)) #bf_tda_pca_5.50.5_rf.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n4_test))
##Node5
DryBean_TDA_PC_5.50.5_n5_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning: model fit failed for Fold2: mtry= 50 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 100 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 150 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 200 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 250 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 300 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 350 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 400 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 450 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 500 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 550 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 600 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 650 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 700 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 750 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 800 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 850 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 900 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry= 950 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning: model fit failed for Fold2: mtry=1000 Error in randomForest.default(x, y, mtry = param$mtry, ...) :
## Need at least two classes to do classification.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_PC_5.50.5_n5_RfFit0
## Random Forest
##
## 417 samples
## 16 predictor
## 2 classes: 'BOMBAY', 'CALI'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 278, 278, 278
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 1 NaN
## 100 1 NaN
## 150 1 NaN
## 200 1 NaN
## 250 1 NaN
## 300 1 NaN
## 350 1 NaN
## 400 1 NaN
## 450 1 NaN
## 500 1 NaN
## 550 1 NaN
## 600 1 NaN
## 650 1 NaN
## 700 1 NaN
## 750 1 NaN
## 800 1 NaN
## 850 1 NaN
## 900 1 NaN
## 950 1 NaN
## 1000 1 NaN
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 50.
DryBean_TDA_PC_5.50.5_n5_RfFit0$resample
## Accuracy Kappa Resample
## 1 1 NA Fold1
## 2 1 NA Fold3
## 3 NA NA Fold2
db_tda_pc_5.50.5_n5_rf_fit0_re<-DryBean_TDA_PC_5.50.5_n5_RfFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n5_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 417 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 834 matrix numeric
## oob.times 417 -none- numeric
## classes 2 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 417 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## param 1 -none- list
vip(DryBean_TDA_PC_5.50.5_n5_RfFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n5_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_PC_5.50.5_n5_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n5_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n5_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n5_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 1 156 0 0 0 0 0
## CALI 395 0 489 1063 578 608 790
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.1581
## 95% CI : (0.147, 0.1696)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0468
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 1.00000 0.0000
## Specificity 1.00000 0.99975 0.04372 1.0000
## Pos Pred Value NaN 0.99363 0.12465 NaN
## Neg Pred Value 0.90294 1.00000 1.00000 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.00000 0.03824 0.11985 0.0000
## Detection Prevalence 0.00000 0.03848 0.96152 0.0000
## Balanced Accuracy 0.50000 0.99987 0.52186 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 1 156 0 0 0 0 0
## CALI 395 0 489 1063 578 608 790
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.1581
## 95% CI : (0.147, 0.1696)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0468
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 1.00000 0.0000
## Specificity 1.00000 0.99975 0.04372 1.0000
## Pos Pred Value NaN 0.99363 0.12465 NaN
## Neg Pred Value 0.90294 1.00000 1.00000 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.00000 0.03824 0.11985 0.0000
## Detection Prevalence 0.00000 0.03848 0.96152 0.0000
## Balanced Accuracy 0.50000 0.99987 0.52186 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.15808824 0.04684314 0.14701909 0.16964931 0.26053922
## AccuracyPValue McnemarPValue
## 1.00000000 NaN
db_tda_pc_5.50.5_n5_rf_cf0_ov_acc<-db_tda_pc_5.50.5_n5_rf_cf0$overall[1]
db_tda_pc_5.50.5_n5_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0 1.00000000 NaN 0.9029412 NA
## Class: BOMBAY 1 0.99974516 0.9936306 1.0000000 0.9936306
## Class: CALI 1 0.04372041 0.1246495 1.0000000 0.1246495
## Class: DERMASON 0 1.00000000 NaN 0.7394608 NA
## Class: HOROZ 0 1.00000000 NaN 0.8583333 NA
## Class: SEKER 0 1.00000000 NaN 0.8509804 NA
## Class: SIRA 0 1.00000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate Detection Prevalence
## Class: BARBUNYA 0 NA 0.09705882 0.00000000 0.00000000
## Class: BOMBAY 1 0.9968051 0.03823529 0.03823529 0.03848039
## Class: CALI 1 0.2216682 0.11985294 0.11985294 0.96151961
## Class: DERMASON 0 NA 0.26053922 0.00000000 0.00000000
## Class: HOROZ 0 NA 0.14166667 0.00000000 0.00000000
## Class: SEKER 0 NA 0.14901961 0.00000000 0.00000000
## Class: SIRA 0 NA 0.19362745 0.00000000 0.00000000
## Balanced Accuracy
## Class: BARBUNYA 0.5000000
## Class: BOMBAY 0.9998726
## Class: CALI 0.5218602
## Class: DERMASON 0.5000000
## Class: HOROZ 0.5000000
## Class: SEKER 0.5000000
## Class: SIRA 0.5000000
db_tda_pc_5.50.5_n5_rf_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n5_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_rf_n5_3_fold<-(db_rf_fit_re-db_tda_pc_5.50.5_n5_rf_fit0_re)
diff_drybean_tda_pca_5.50.5_rf_n5_3_fold
## Accuracy
## 1 -0.07491344
## 2 -0.07614852
## 3 NA
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold
## $probLeft
## [1] NA
##
## $probRope
## [1] NA
##
## $probRight
## [1] NA
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n5_3_fold_odds.left
## [1] NA
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_rf.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_rf.n5_3_fold
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n5_3_fold
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold))
#bf_tda_pca_5.50.5_rf.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_rf_n5_3_fold)
## t = -122.31, df = 1, p-value = 0.005205
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.08337758 -0.06768439
## sample estimates:
## mean of x
## -0.07553098
### Test set diff
diff_drybean_tda_pca_5.50.5_rf.n5_test<-(db_rf_cf_ov_acc-db_tda_pc_5.50.5_n5_rf_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_rf.n5_test
## Accuracy
## 0.7598039
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n5_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_rf.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_rf.n5_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_rf.n5_test$probLeft/bst_dbf_db_tda_pca_5.50.5_rf.n5_test$probRight
bst_dbf_db_tda_pca_5.50.5_rf.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_rf.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n5_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_rf.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1565333
##
## $winRight
## [1] 0.8434667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_rf.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n5_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_rf.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_rf.n5_test))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_rf.n5_test)) #bf_tda_pca_5.50.5_rf.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_rf.n5_test))
##With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_KDE_5.50.5_n1_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_KDE_5.50.5_n1_RfFit0
## Random Forest
##
## 8473 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5649, 5648, 5649
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9512562 0.9417543
## 100 0.9511381 0.9416175
## 150 0.9507841 0.9411931
## 200 0.9512562 0.9417534
## 250 0.9505481 0.9409065
## 300 0.9511382 0.9416147
## 350 0.9509022 0.9413309
## 400 0.9509022 0.9413307
## 450 0.9503121 0.9406302
## 500 0.9507843 0.9411942
## 550 0.9512562 0.9417537
## 600 0.9514924 0.9420387
## 650 0.9505481 0.9409120
## 700 0.9513743 0.9418963
## 750 0.9512564 0.9417572
## 800 0.9510201 0.9414767
## 850 0.9507843 0.9411931
## 900 0.9512564 0.9417549
## 950 0.9511382 0.9416140
## 1000 0.9510202 0.9414757
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 600.
DryBean_TDA_KDE_5.50.5_n1_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9564602 0.9479760 Fold2
## 2 0.9514873 0.9420608 Fold1
## 3 0.9465297 0.9360791 Fold3
ad_tda_kde_5.50.5_n1_rf_fit0_re<-DryBean_TDA_KDE_5.50.5_n1_RfFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n1_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 8473 factor numeric
## err.rate 4000 -none- numeric
## confusion 56 -none- numeric
## votes 59311 matrix numeric
## oob.times 8473 -none- numeric
## classes 7 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 8473 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 7 -none- character
## param 1 -none- list
vip(DryBean_TDA_KDE_5.50.5_n1_RfFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n1_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n1_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n1_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n1_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
ad_tda_kde_5.50.5_n1_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 394 0 0 1 0 0 1
## BOMBAY 0 156 0 0 0 0 0
## CALI 0 0 488 0 0 0 0
## DERMASON 1 0 0 881 6 3 119
## HOROZ 0 0 0 1 572 0 0
## SEKER 0 0 0 32 0 589 11
## SIRA 1 0 1 148 0 16 659
##
## Overall Statistics
##
## Accuracy : 0.9164
## 95% CI : (0.9075, 0.9247)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8991
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.99495 1.00000 0.9980 0.8288
## Specificity 0.99946 1.00000 1.0000 0.9572
## Pos Pred Value 0.99495 1.00000 1.0000 0.8723
## Neg Pred Value 0.99946 1.00000 0.9997 0.9407
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09657 0.03824 0.1196 0.2159
## Detection Prevalence 0.09706 0.03824 0.1196 0.2475
## Balanced Accuracy 0.99720 1.00000 0.9990 0.8930
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9896 0.9688 0.8342
## Specificity 0.9997 0.9876 0.9495
## Pos Pred Value 0.9983 0.9320 0.7988
## Neg Pred Value 0.9983 0.9945 0.9598
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1402 0.1444 0.1615
## Detection Prevalence 0.1404 0.1549 0.2022
## Balanced Accuracy 0.9947 0.9782 0.8919
ad_tda_kde_5.50.5_n1_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 394 0 0 1 0 0 1
## BOMBAY 0 156 0 0 0 0 0
## CALI 0 0 488 0 0 0 0
## DERMASON 1 0 0 881 6 3 119
## HOROZ 0 0 0 1 572 0 0
## SEKER 0 0 0 32 0 589 11
## SIRA 1 0 1 148 0 16 659
##
## Overall Statistics
##
## Accuracy : 0.9164
## 95% CI : (0.9075, 0.9247)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8991
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.99495 1.00000 0.9980 0.8288
## Specificity 0.99946 1.00000 1.0000 0.9572
## Pos Pred Value 0.99495 1.00000 1.0000 0.8723
## Neg Pred Value 0.99946 1.00000 0.9997 0.9407
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09657 0.03824 0.1196 0.2159
## Detection Prevalence 0.09706 0.03824 0.1196 0.2475
## Balanced Accuracy 0.99720 1.00000 0.9990 0.8930
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9896 0.9688 0.8342
## Specificity 0.9997 0.9876 0.9495
## Pos Pred Value 0.9983 0.9320 0.7988
## Neg Pred Value 0.9983 0.9945 0.9598
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1402 0.1444 0.1615
## Detection Prevalence 0.1404 0.1549 0.2022
## Balanced Accuracy 0.9947 0.9782 0.8919
ad_tda_kde_5.50.5_n1_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9164216 0.8990787 0.9075056 0.9247357 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n1_rf_cf0_ov_acc<-ad_tda_kde_5.50.5_n1_rf_cf0$overall[1]
ad_tda_kde_5.50.5_n1_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9949495 0.9994571 0.9949495 0.9994571 0.9949495
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9979550 1.0000000 1.0000000 0.9997216 1.0000000
## Class: DERMASON 0.8287865 0.9572423 0.8722772 0.9407166 0.8722772
## Class: HOROZ 0.9896194 0.9997144 0.9982548 0.9982891 0.9982548
## Class: SEKER 0.9687500 0.9876152 0.9319620 0.9944896 0.9319620
## Class: SIRA 0.8341772 0.9495441 0.7987879 0.9597542 0.7987879
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9949495 0.9949495 0.09705882 0.09656863
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9979550 0.9989765 0.11985294 0.11960784
## Class: DERMASON 0.8287865 0.8499759 0.26053922 0.21593137
## Class: HOROZ 0.9896194 0.9939183 0.14166667 0.14019608
## Class: SEKER 0.9687500 0.9500000 0.14901961 0.14436275
## Class: SIRA 0.8341772 0.8160991 0.19362745 0.16151961
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09705882 0.9972033
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.11960784 0.9989775
## Class: DERMASON 0.24754902 0.8930144
## Class: HOROZ 0.14044118 0.9946669
## Class: SEKER 0.15490196 0.9781826
## Class: SIRA 0.20220588 0.8918606
ad_tda_kde_5.50.5_n1_rf_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n1_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_rf_n1_3_fold<-(db_rf_fit_re-ad_tda_kde_5.50.5_n1_rf_fit0_re)
diff_drybean_tda_kde_5.50.5_rf_n1_3_fold
## Accuracy
## 1 -0.03137362
## 2 -0.02763577
## 3 -0.01579927
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n1_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n1_3_fold_odds.left<-bst_tda_kde_5.50.5_rf.n1_3_fold$probLeft/bst_tda_kde_5.50.5_rf.n1_3_fold$probRight
bst_tda_kde_5.50.5_rf.n1_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n1_3_fold
## $winLeft
## [1] 0.9629333
##
## $winRope
## [1] 0.03706667
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n1_3_fold
## $left
## [1] 0.9448367
##
## $rope
## [1] 0.04354533
##
## $right
## [1] 0.01161796
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold))
#bf_tda_kde_5.50.5_rf.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_rf_n1_3_fold)
## t = -5.3122, df = 2, p-value = 0.03366
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.04513362 -0.00473882
## sample estimates:
## mean of x
## -0.02493622
### Test set diff
diff_drybean_tda_kde_5.50.5_rf.n1_test<-(db_rf_cf_ov_acc-ad_tda_kde_5.50.5_n1_rf_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_rf.n1_test
## Accuracy
## 0.001470588
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n1_test),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 1
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n1_test_odds.left<-bst_tda_kde_5.50.5_rf.n1_test$probLeft/bst_tda_kde_5.50.5_rf.n1_test$probRight
bst_tda_kde_5.50.5_rf.n1_test_odds.left
## [1] NaN
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n1_test),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 1
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n1_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_rf.n1_test))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf.n1_test)) #bf_tda_kde_5.50.5_rf.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n1_test))
##Node2
DryBean_TDA_KDE_5.50.5_n2_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_KDE_5.50.5_n2_RfFit0
## Random Forest
##
## 7582 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5054, 5055, 5055
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9444737 0.9290609
## 100 0.9422315 0.9261804
## 150 0.9436822 0.9280384
## 200 0.9430229 0.9271944
## 250 0.9435506 0.9278785
## 300 0.9430231 0.9271971
## 350 0.9431549 0.9273653
## 400 0.9435505 0.9278756
## 450 0.9419678 0.9258586
## 500 0.9430229 0.9271926
## 550 0.9424954 0.9265167
## 600 0.9430228 0.9272057
## 650 0.9428913 0.9270361
## 700 0.9419678 0.9258475
## 750 0.9432868 0.9275378
## 800 0.9423634 0.9263604
## 850 0.9424954 0.9265269
## 900 0.9439459 0.9283897
## 950 0.9420998 0.9260186
## 1000 0.9418359 0.9256763
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 50.
DryBean_TDA_KDE_5.50.5_n2_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9450158 0.9298876 Fold1
## 2 0.9457855 0.9306419 Fold3
## 3 0.9426197 0.9266532 Fold2
ad_tda_KDE_5.50.5_n2_rf_fit0_re<-DryBean_TDA_KDE_5.50.5_n2_RfFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n2_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 7582 factor numeric
## err.rate 3500 -none- numeric
## confusion 42 -none- numeric
## votes 45492 matrix numeric
## oob.times 7582 -none- numeric
## classes 6 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 7582 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 6 -none- character
## param 1 -none- list
vip(DryBean_TDA_KDE_5.50.5_n2_RfFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n2_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n2_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n2_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n2_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n2_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 340 1 5 0 3 1 1
## BOMBAY 0 0 0 0 0 0 0
## CALI 10 150 475 0 13 0 3
## DERMASON 0 0 0 1007 3 9 96
## HOROZ 3 0 6 1 558 0 1
## SEKER 7 0 0 23 0 593 3
## SIRA 36 5 3 32 1 5 686
##
## Overall Statistics
##
## Accuracy : 0.8968
## 95% CI : (0.8871, 0.906)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8745
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.85859 0.00000 0.9714 0.9473
## Specificity 0.99701 1.00000 0.9510 0.9642
## Pos Pred Value 0.96866 NaN 0.7296 0.9031
## Neg Pred Value 0.98498 0.96176 0.9959 0.9811
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08333 0.00000 0.1164 0.2468
## Detection Prevalence 0.08603 0.00000 0.1596 0.2733
## Balanced Accuracy 0.92780 0.50000 0.9612 0.9558
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9654 0.9753 0.8684
## Specificity 0.9969 0.9905 0.9751
## Pos Pred Value 0.9807 0.9473 0.8932
## Neg Pred Value 0.9943 0.9957 0.9686
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1368 0.1453 0.1681
## Detection Prevalence 0.1395 0.1534 0.1882
## Balanced Accuracy 0.9811 0.9829 0.9217
ad_tda_kde_5.50.5_n2_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 340 1 5 0 3 1 1
## BOMBAY 0 0 0 0 0 0 0
## CALI 10 150 475 0 13 0 3
## DERMASON 0 0 0 1007 3 9 96
## HOROZ 3 0 6 1 558 0 1
## SEKER 7 0 0 23 0 593 3
## SIRA 36 5 3 32 1 5 686
##
## Overall Statistics
##
## Accuracy : 0.8968
## 95% CI : (0.8871, 0.906)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8745
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.85859 0.00000 0.9714 0.9473
## Specificity 0.99701 1.00000 0.9510 0.9642
## Pos Pred Value 0.96866 NaN 0.7296 0.9031
## Neg Pred Value 0.98498 0.96176 0.9959 0.9811
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08333 0.00000 0.1164 0.2468
## Detection Prevalence 0.08603 0.00000 0.1596 0.2733
## Balanced Accuracy 0.92780 0.50000 0.9612 0.9558
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9654 0.9753 0.8684
## Specificity 0.9969 0.9905 0.9751
## Pos Pred Value 0.9807 0.9473 0.8932
## Neg Pred Value 0.9943 0.9957 0.9686
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1368 0.1453 0.1681
## Detection Prevalence 0.1395 0.1534 0.1882
## Balanced Accuracy 0.9811 0.9829 0.9217
ad_tda_kde_5.50.5_n2_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8968137 0.8745084 0.8870710 0.9059839 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n2_rf_cf0_ov_acc<-ad_tda_kde_5.50.5_n2_rf_cf0$overall[1]
ad_tda_kde_5.50.5_n2_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.8585859 0.9970141 0.9686610 0.9849826 0.9686610
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.9713701 0.9509886 0.7296467 0.9959172 0.7296467
## Class: DERMASON 0.9473189 0.9642029 0.9031390 0.9811130 0.9031390
## Class: HOROZ 0.9653979 0.9968589 0.9806678 0.9943036 0.9806678
## Class: SEKER 0.9753289 0.9904954 0.9472843 0.9956572 0.9472843
## Class: SIRA 0.8683544 0.9750760 0.8932292 0.9685990 0.8932292
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.8585859 0.9103079 0.09705882 0.08333333
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.9713701 0.8333333 0.11985294 0.11642157
## Class: DERMASON 0.9473189 0.9247016 0.26053922 0.24681373
## Class: HOROZ 0.9653979 0.9729730 0.14166667 0.13676471
## Class: SEKER 0.9753289 0.9611021 0.14901961 0.14534314
## Class: SIRA 0.8683544 0.8806162 0.19362745 0.16813725
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.08602941 0.9278000
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.15955882 0.9611794
## Class: DERMASON 0.27328431 0.9557609
## Class: HOROZ 0.13946078 0.9811284
## Class: SEKER 0.15343137 0.9829122
## Class: SIRA 0.18823529 0.9217152
ad_tda_kde_5.50.5_n2_rf_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n2_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_rf_n2_3_fold<-(db_rf_fit_re-ad_tda_KDE_5.50.5_n2_rf_fit0_re)
diff_drybean_tda_kde_5.50.5_rf_n2_3_fold
## Accuracy
## 1 -0.01992926
## 2 -0.02193404
## 3 -0.01188923
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n2_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n2_3_fold_odds.left<-bst_tda_kde_5.50.5_rf.n2_3_fold$probLeft/bst_tda_kde_5.50.5_rf.n2_3_fold$probRight
bst_tda_kde_5.50.5_rf.n2_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n2_3_fold
## $winLeft
## [1] 0.9129667
##
## $winRope
## [1] 0.08703333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n2_3_fold
## $left
## [1] 0.9224685
##
## $rope
## [1] 0.0696636
##
## $right
## [1] 0.007867899
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold))
#bf_tda_kde_5.50.5_rf.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_rf_n2_3_fold)
## t = -5.8378, df = 2, p-value = 0.02811
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.031123199 -0.004711821
## sample estimates:
## mean of x
## -0.01791751
### Test set diff
diff_drybean_tda_kde_5.50.5_rf.n2_test<-(db_rf_cf_ov_acc-ad_tda_kde_5.50.5_n2_rf_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_rf.n2_test
## Accuracy
## 0.02107843
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n2_test),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n2_test_odds.left<-bst_tda_kde_5.50.5_rf.n2_test$probLeft/bst_tda_kde_5.50.5_rf.n2_test$probRight
bst_tda_kde_5.50.5_rf.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n2_test),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1580667
##
## $winRight
## [1] 0.8419333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n2_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_rf.n2_test))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf.n2_test)) #bf_tda_kde_5.50.5_rf.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n2_test))
##Node3
DryBean_TDA_KDE_5.50.5_n3_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_KDE_5.50.5_n3_RfFit0
## Random Forest
##
## 4149 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 2766, 2766, 2766
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.9103398 0.8647137
## 100 0.9117860 0.8669165
## 150 0.9113039 0.8661831
## 200 0.9103398 0.8647139
## 250 0.9132321 0.8690892
## 300 0.9108219 0.8654454
## 350 0.9108219 0.8654660
## 400 0.9127501 0.8683567
## 450 0.9110629 0.8658288
## 500 0.9117860 0.8669082
## 550 0.9113039 0.8661768
## 600 0.9108219 0.8654641
## 650 0.9113039 0.8661911
## 700 0.9125090 0.8680099
## 750 0.9105809 0.8650758
## 800 0.9129911 0.8687300
## 850 0.9113039 0.8661818
## 900 0.9108219 0.8654443
## 950 0.9122680 0.8676217
## 1000 0.9110629 0.8658204
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 250.
DryBean_TDA_KDE_5.50.5_n3_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.9096168 0.8638074 Fold1
## 2 0.9146782 0.8709429 Fold3
## 3 0.9154013 0.8725173 Fold2
ad_tda_kde_5.50.5_n3_rf_fit0_re<-DryBean_TDA_KDE_5.50.5_n3_RfFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n3_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 4149 factor numeric
## err.rate 3500 -none- numeric
## confusion 42 -none- numeric
## votes 24894 matrix numeric
## oob.times 4149 -none- numeric
## classes 6 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 4149 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 6 -none- character
## param 1 -none- list
vip(DryBean_TDA_KDE_5.50.5_n3_RfFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n3_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n3_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n3_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n3_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n3_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 297 64 129 0 9 1 15
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 1 0 0 0 0
## DERMASON 0 0 0 1032 6 5 61
## HOROZ 22 18 259 0 553 0 6
## SEKER 16 4 1 15 0 588 3
## SIRA 61 70 99 16 10 14 705
##
## Overall Statistics
##
## Accuracy : 0.7784
## 95% CI : (0.7654, 0.7911)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.7292
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.75000 0.00000 0.0020450 0.9708
## Specificity 0.94083 1.00000 1.0000000 0.9761
## Pos Pred Value 0.57670 NaN 1.0000000 0.9348
## Neg Pred Value 0.97223 0.96176 0.8803628 0.9896
## Prevalence 0.09706 0.03824 0.1198529 0.2605
## Detection Rate 0.07279 0.00000 0.0002451 0.2529
## Detection Prevalence 0.12623 0.00000 0.0002451 0.2706
## Balanced Accuracy 0.84541 0.50000 0.5010225 0.9735
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9671 0.8924
## Specificity 0.9129 0.9888 0.9179
## Pos Pred Value 0.6445 0.9378 0.7231
## Neg Pred Value 0.9922 0.9942 0.9726
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1441 0.1728
## Detection Prevalence 0.2103 0.1537 0.2390
## Balanced Accuracy 0.9348 0.9779 0.9052
ad_tda_kde_5.50.5_n3_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 297 64 129 0 9 1 15
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 1 0 0 0 0
## DERMASON 0 0 0 1032 6 5 61
## HOROZ 22 18 259 0 553 0 6
## SEKER 16 4 1 15 0 588 3
## SIRA 61 70 99 16 10 14 705
##
## Overall Statistics
##
## Accuracy : 0.7784
## 95% CI : (0.7654, 0.7911)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.7292
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.75000 0.00000 0.0020450 0.9708
## Specificity 0.94083 1.00000 1.0000000 0.9761
## Pos Pred Value 0.57670 NaN 1.0000000 0.9348
## Neg Pred Value 0.97223 0.96176 0.8803628 0.9896
## Prevalence 0.09706 0.03824 0.1198529 0.2605
## Detection Rate 0.07279 0.00000 0.0002451 0.2529
## Detection Prevalence 0.12623 0.00000 0.0002451 0.2706
## Balanced Accuracy 0.84541 0.50000 0.5010225 0.9735
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9671 0.8924
## Specificity 0.9129 0.9888 0.9179
## Pos Pred Value 0.6445 0.9378 0.7231
## Neg Pred Value 0.9922 0.9942 0.9726
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1441 0.1728
## Detection Prevalence 0.2103 0.1537 0.2390
## Balanced Accuracy 0.9348 0.9779 0.9052
ad_tda_kde_5.50.5_n3_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.7784314 0.7292187 0.7653685 0.7910943 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n3_rf_cf0_ov_acc<-ad_tda_kde_5.50.5_n3_rf_cf0$overall[1]
ad_tda_kde_5.50.5_n3_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.75000000 0.9408252 0.5766990 0.9722300 0.5766990
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.00204499 1.0000000 1.0000000 0.8803628 1.0000000
## Class: DERMASON 0.97083725 0.9761352 0.9347826 0.9895833 0.9347826
## Class: HOROZ 0.95674740 0.9129069 0.6445221 0.9922408 0.6445221
## Class: SEKER 0.96710526 0.9887673 0.9377990 0.9942079 0.9377990
## Class: SIRA 0.89240506 0.9179331 0.7230769 0.9726248 0.7230769
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.75000000 0.652030735 0.09705882 0.072794118
## Class: BOMBAY 0.00000000 NA 0.03823529 0.000000000
## Class: CALI 0.00204499 0.004081633 0.11985294 0.000245098
## Class: DERMASON 0.97083725 0.952468851 0.26053922 0.252941176
## Class: HOROZ 0.95674740 0.770194986 0.14166667 0.135539216
## Class: SEKER 0.96710526 0.952226721 0.14901961 0.144117647
## Class: SIRA 0.89240506 0.798866856 0.19362745 0.172794118
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.126225490 0.8454126
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000245098 0.5010225
## Class: DERMASON 0.270588235 0.9734862
## Class: HOROZ 0.210294118 0.9348272
## Class: SEKER 0.153676471 0.9779363
## Class: SIRA 0.238970588 0.9051691
ad_tda_kde_5.50.5_n3_rf_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n3_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_rf_n3_3_fold<-(db_rf_fit_re-ad_tda_kde_5.50.5_n3_rf_fit0_re)
diff_drybean_tda_kde_5.50.5_rf_n3_3_fold
## Accuracy
## 1 0.015469785
## 2 0.009173243
## 3 0.015329177
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n3_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n3_3_fold_odds.left<-bst_tda_kde_5.50.5_rf.n3_3_fold$probLeft/bst_tda_kde_5.50.5_rf.n3_3_fold$probRight
bst_tda_kde_5.50.5_rf.n3_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n3_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.3018667
##
## $winRight
## [1] 0.6981333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n3_3_fold
## $left
## [1] 0.005198295
##
## $rope
## [1] 0.1447256
##
## $right
## [1] 0.8500761
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold))
#bf_tda_kde_5.50.5_rf.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_rf_n3_3_fold)
## t = 6.4187, df = 2, p-value = 0.02342
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.004392581 0.022255555
## sample estimates:
## mean of x
## 0.01332407
### Test set diff
diff_drybean_tda_kde_5.50.5_rf.n3_test<-(db_rf_cf_ov_acc-ad_tda_kde_5.50.5_n3_rf_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_rf.n3_test
## Accuracy
## 0.1394608
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n3_test),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n3_test_odds.left<-bst_tda_kde_5.50.5_rf.n3_test$probLeft/bst_tda_kde_5.50.5_rf.n3_test$probRight
bst_tda_kde_5.50.5_rf.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n3_test),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1580333
##
## $winRight
## [1] 0.8419667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n3_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_rf.n3_test))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf.n3_test)) #bf_tda_kde_5.50.5_rf.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n3_test))
##Node4
DryBean_TDA_KDE_5.50.5_n4_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_KDE_5.50.5_n4_RfFit0
## Random Forest
##
## 2024 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1348, 1351, 1349
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.8196610 0.6997197
## 100 0.8152114 0.6921414
## 150 0.8176879 0.6958127
## 200 0.8171904 0.6949336
## 250 0.8171852 0.6960052
## 300 0.8186667 0.6985042
## 350 0.8147198 0.6919284
## 400 0.8157082 0.6934125
## 450 0.8157067 0.6929295
## 500 0.8166995 0.6948643
## 550 0.8147183 0.6915713
## 600 0.8201460 0.7003374
## 650 0.8181707 0.6973689
## 700 0.8176813 0.6966316
## 750 0.8132354 0.6891959
## 800 0.8166899 0.6956077
## 850 0.8196632 0.6995717
## 900 0.8176871 0.6959447
## 950 0.8191664 0.6990904
## 1000 0.8157060 0.6927469
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 600.
DryBean_TDA_KDE_5.50.5_n4_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.8142645 0.6923974 Fold2
## 2 0.8328402 0.7210895 Fold1
## 3 0.8133333 0.6875253 Fold3
ad_tda_kde_5.50.5_n4_rf_fit0_re<-DryBean_TDA_KDE_5.50.5_n4_RfFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n4_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 2024 factor numeric
## err.rate 2500 -none- numeric
## confusion 20 -none- numeric
## votes 8096 matrix numeric
## oob.times 2024 -none- numeric
## classes 4 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 2024 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 4 -none- character
## param 1 -none- list
vip(DryBean_TDA_KDE_5.50.5_n4_RfFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n4_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n4_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n4_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n4_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n4_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 326 75 342 1015 491 39 337
## HOROZ 0 0 0 0 3 0 0
## SEKER 24 5 0 9 0 561 6
## SIRA 46 76 147 39 84 8 447
##
## Overall Statistics
##
## Accuracy : 0.4966
## 95% CI : (0.4811, 0.512)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3462
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9548
## Specificity 1.00000 1.00000 1.0000 0.4664
## Pos Pred Value NaN NaN NaN 0.3867
## Neg Pred Value 0.90294 0.96176 0.8801 0.9670
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2488
## Detection Prevalence 0.00000 0.00000 0.0000 0.6434
## Balanced Accuracy 0.50000 0.50000 0.5000 0.7106
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0051903 0.9227 0.5658
## Specificity 1.0000000 0.9873 0.8784
## Pos Pred Value 1.0000000 0.9273 0.5277
## Neg Pred Value 0.8589649 0.9865 0.8939
## Prevalence 0.1416667 0.1490 0.1936
## Detection Rate 0.0007353 0.1375 0.1096
## Detection Prevalence 0.0007353 0.1483 0.2076
## Balanced Accuracy 0.5025952 0.9550 0.7221
ad_tda_kde_5.50.5_n4_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 326 75 342 1015 491 39 337
## HOROZ 0 0 0 0 3 0 0
## SEKER 24 5 0 9 0 561 6
## SIRA 46 76 147 39 84 8 447
##
## Overall Statistics
##
## Accuracy : 0.4966
## 95% CI : (0.4811, 0.512)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3462
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9548
## Specificity 1.00000 1.00000 1.0000 0.4664
## Pos Pred Value NaN NaN NaN 0.3867
## Neg Pred Value 0.90294 0.96176 0.8801 0.9670
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2488
## Detection Prevalence 0.00000 0.00000 0.0000 0.6434
## Balanced Accuracy 0.50000 0.50000 0.5000 0.7106
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0051903 0.9227 0.5658
## Specificity 1.0000000 0.9873 0.8784
## Pos Pred Value 1.0000000 0.9273 0.5277
## Neg Pred Value 0.8589649 0.9865 0.8939
## Prevalence 0.1416667 0.1490 0.1936
## Detection Rate 0.0007353 0.1375 0.1096
## Detection Prevalence 0.0007353 0.1483 0.2076
## Balanced Accuracy 0.5025952 0.9550 0.7221
ad_tda_kde_5.50.5_n4_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.965686e-01 3.461723e-01 4.811113e-01 5.120309e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 5.748350e-227 NaN
ad_tda_kde_5.50.5_n4_rf_cf0_ov_acc<-ad_tda_kde_5.50.5_n4_rf_cf0$overall[1]
ad_tda_kde_5.50.5_n4_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.000000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.000000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.000000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.954844779 0.4663573 0.3866667 0.9670103 0.3866667
## Class: HOROZ 0.005190311 1.0000000 1.0000000 0.8589649 1.0000000
## Class: SEKER 0.922697368 0.9873272 0.9272727 0.9864748 0.9272727
## Class: SIRA 0.565822785 0.8784195 0.5277450 0.8939066 0.5277450
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.000000000 NA 0.09705882 0.0000000000
## Class: BOMBAY 0.000000000 NA 0.03823529 0.0000000000
## Class: CALI 0.000000000 NA 0.11985294 0.0000000000
## Class: DERMASON 0.954844779 0.55043384 0.26053922 0.2487745098
## Class: HOROZ 0.005190311 0.01032702 0.14166667 0.0007352941
## Class: SEKER 0.922697368 0.92497939 0.14901961 0.1375000000
## Class: SIRA 0.565822785 0.54612095 0.19362745 0.1095588235
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000000 0.5000000
## Class: BOMBAY 0.0000000000 0.5000000
## Class: CALI 0.0000000000 0.5000000
## Class: DERMASON 0.6433823529 0.7106010
## Class: HOROZ 0.0007352941 0.5025952
## Class: SEKER 0.1482843137 0.9550123
## Class: SIRA 0.2075980392 0.7221211
ad_tda_kde_5.50.5_n4_rf_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n4_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_rf_n4_3_fold<-(db_rf_fit_re-ad_tda_kde_5.50.5_n4_rf_fit0_re)
diff_drybean_tda_kde_5.50.5_rf_n4_3_fold
## Accuracy
## 1 0.11082207
## 2 0.09101124
## 3 0.11739715
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n4_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n4_3_fold_odds.left<-bst_tda_kde_5.50.5_rf.n4_3_fold$probLeft/bst_tda_kde_5.50.5_rf.n4_3_fold$probRight
bst_tda_kde_5.50.5_rf.n4_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n4_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.008733333
##
## $winRight
## [1] 0.9912667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n4_3_fold
## $left
## [1] 0.003065215
##
## $rope
## [1] 0.001384956
##
## $right
## [1] 0.9955498
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold,c(-0.01,0.01)))
### Test set diff
diff_drybean_tda_kde_5.50.5_rf.n4_test<-(db_rf_cf_ov_acc-ad_tda_kde_5.50.5_n4_rf_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_rf.n4_test
## Accuracy
## 0.4213235
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n4_test),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
#BayesFactor
#bf_tda_kde_5.50.5_rf.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold))
#bf_tda_kde_5.50.5_rf.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_rf_n4_3_fold)
## t = 13.419, df = 2, p-value = 0.005508
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.07229029 0.14053001
## sample estimates:
## mean of x
## 0.1064102
bst_tda_kde_5.50.5_rf.n4_test_odds.left<-bst_tda_kde_5.50.5_rf.n4_test$probLeft/bst_tda_kde_5.50.5_rf.n4_test$probRight
bst_tda_kde_5.50.5_rf.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1589667
##
## $winRight
## [1] 0.8410333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n4_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_rf.n4_test))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf.n4_test)) #bf_tda_kde_5.50.5_rf.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n4_test))
##Node5
DryBean_TDA_KDE_5.50.5_n5_RfFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'rf',
trControl = fitControl,
tuneGrid = rfGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
## Warning in randomForest.default(x, y, mtry = param$mtry, ...): invalid mtry:
## reset to within valid range
DryBean_TDA_KDE_5.50.5_n5_RfFit0
## Random Forest
##
## 989 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 660, 658, 660
## Resampling results across tuning parameters:
##
## mtry Accuracy Kappa
## 50 0.7259693 0.5224857
## 100 0.7239307 0.5212230
## 150 0.7279651 0.5258962
## 200 0.7148061 0.5022183
## 250 0.7249255 0.5197732
## 300 0.7229114 0.5173059
## 350 0.7269580 0.5245006
## 400 0.7269580 0.5246805
## 450 0.7259510 0.5214657
## 500 0.7269764 0.5240011
## 550 0.7259387 0.5223178
## 600 0.7249378 0.5199934
## 650 0.7229237 0.5190916
## 700 0.7289782 0.5279222
## 750 0.7198719 0.5136756
## 800 0.7259693 0.5215907
## 850 0.7239430 0.5186947
## 900 0.7209035 0.5142262
## 950 0.7198781 0.5148667
## 1000 0.7279589 0.5266978
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was mtry = 700.
DryBean_TDA_KDE_5.50.5_n5_RfFit0$resample
## Accuracy Kappa Resample
## 1 0.7173252 0.5220190 Fold1
## 2 0.7203647 0.4970255 Fold3
## 3 0.7492447 0.5647221 Fold2
ad_tda_kde_5.50.5_n5_rf_fit0_re<-DryBean_TDA_KDE_5.50.5_n5_RfFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n5_RfFit0)
## Length Class Mode
## call 5 -none- call
## type 1 -none- character
## predicted 989 factor numeric
## err.rate 2500 -none- numeric
## confusion 20 -none- numeric
## votes 3956 matrix numeric
## oob.times 989 -none- numeric
## classes 4 -none- character
## importance 16 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 989 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 4 -none- character
## param 1 -none- list
vip(DryBean_TDA_KDE_5.50.5_n5_RfFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n5_RfFit TDA-Assited RF")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n5_RfFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n5_RfFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n5_rf_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n5_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 111 7 16 920 14 35 136
## HOROZ 0 0 0 0 1 0 0
## SEKER 12 4 0 18 0 562 9
## SIRA 273 145 473 125 563 11 645
##
## Overall Statistics
##
## Accuracy : 0.5216
## 95% CI : (0.5061, 0.537)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3964
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.8655
## Specificity 1.00000 1.00000 1.0000 0.8943
## Pos Pred Value NaN NaN NaN 0.7425
## Neg Pred Value 0.90294 0.96176 0.8801 0.9497
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2255
## Detection Prevalence 0.00000 0.00000 0.0000 0.3037
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8799
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0017301 0.9243 0.8165
## Specificity 1.0000000 0.9876 0.5167
## Pos Pred Value 1.0000000 0.9289 0.2886
## Neg Pred Value 0.8585438 0.9868 0.9214
## Prevalence 0.1416667 0.1490 0.1936
## Detection Rate 0.0002451 0.1377 0.1581
## Detection Prevalence 0.0002451 0.1483 0.5478
## Balanced Accuracy 0.5008651 0.9560 0.6666
ad_tda_kde_5.50.5_n5_rf_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 111 7 16 920 14 35 136
## HOROZ 0 0 0 0 1 0 0
## SEKER 12 4 0 18 0 562 9
## SIRA 273 145 473 125 563 11 645
##
## Overall Statistics
##
## Accuracy : 0.5216
## 95% CI : (0.5061, 0.537)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3964
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.8655
## Specificity 1.00000 1.00000 1.0000 0.8943
## Pos Pred Value NaN NaN NaN 0.7425
## Neg Pred Value 0.90294 0.96176 0.8801 0.9497
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2255
## Detection Prevalence 0.00000 0.00000 0.0000 0.3037
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8799
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0017301 0.9243 0.8165
## Specificity 1.0000000 0.9876 0.5167
## Pos Pred Value 1.0000000 0.9289 0.2886
## Neg Pred Value 0.8585438 0.9868 0.9214
## Prevalence 0.1416667 0.1490 0.1936
## Detection Rate 0.0002451 0.1377 0.1581
## Detection Prevalence 0.0002451 0.1483 0.5478
## Balanced Accuracy 0.5008651 0.9560 0.6666
ad_tda_kde_5.50.5_n5_rf_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.215686e-01 3.964385e-01 5.061073e-01 5.369990e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 8.315958e-275 NaN
ad_tda_kde_5.50.5_n5_rf_cf0_ov_acc<-ad_tda_kde_5.50.5_n5_rf_cf0$overall[1]
ad_tda_kde_5.50.5_n5_rf_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.000000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.000000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.000000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.865475071 0.8942658 0.7425343 0.9496656 0.7425343
## Class: HOROZ 0.001730104 1.0000000 1.0000000 0.8585438 1.0000000
## Class: SEKER 0.924342105 0.9876152 0.9289256 0.9867626 0.9289256
## Class: SIRA 0.816455696 0.5167173 0.2885906 0.9214092 0.2885906
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.000000000 NA 0.09705882 0.000000000
## Class: BOMBAY 0.000000000 NA 0.03823529 0.000000000
## Class: CALI 0.000000000 NA 0.11985294 0.000000000
## Class: DERMASON 0.865475071 0.799304952 0.26053922 0.225490196
## Class: HOROZ 0.001730104 0.003454231 0.14166667 0.000245098
## Class: SEKER 0.924342105 0.926628195 0.14901961 0.137745098
## Class: SIRA 0.816455696 0.426446281 0.19362745 0.158088235
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.000000000 0.5000000
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000000000 0.5000000
## Class: DERMASON 0.303676471 0.8798704
## Class: HOROZ 0.000245098 0.5008651
## Class: SEKER 0.148284314 0.9559787
## Class: SIRA 0.547794118 0.6665865
ad_tda_kde_5.50.5_n5_rf_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n5_rf_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_rf_n5_3_fold<-(db_rf_fit_re-ad_tda_kde_5.50.5_n5_rf_fit0_re)
diff_drybean_tda_kde_5.50.5_rf_n5_3_fold
## Accuracy
## 1 0.2077613
## 2 0.2034867
## 3 0.1814858
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n5_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n5_3_fold_odds.left<-bst_tda_kde_5.50.5_rf.n5_3_fold$probLeft/bst_tda_kde_5.50.5_rf.n5_3_fold$probRight
bst_tda_kde_5.50.5_rf.n5_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n5_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.008866667
##
## $winRight
## [1] 0.9911333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n5_3_fold
## $left
## [1] 0.001022067
##
## $rope
## [1] 0.0002287087
##
## $right
## [1] 0.9987492
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold))
#bf_tda_kde_5.50.5_rf.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_rf_n5_3_fold)
## t = 24.272, df = 2, p-value = 0.001693
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.1625537 0.2326022
## sample estimates:
## mean of x
## 0.1975779
### Test set diff
diff_drybean_tda_kde_5.50.5_rf.n5_test<-(db_rf_cf_ov_acc-ad_tda_kde_5.50.5_n5_rf_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_rf.n5_test
## Accuracy
## 0.3963235
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n5_test),-0.01,0.01)
bst_tda_kde_5.50.5_rf.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_rf.n5_test_odds.left<-bst_tda_kde_5.50.5_rf.n5_test$probLeft/bst_tda_kde_5.50.5_rf.n5_test$probRight
bst_tda_kde_5.50.5_rf.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_rf.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n5_test),-0.01,0.01)
bsr_tda_kde_5.50.5_rf.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1595667
##
## $winRight
## [1] 0.8404333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_rf.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n5_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_rf.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_rf.n5_test))
#BayesFactor
#bf_tda_kde_5.50.5_rf.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_rf.n5_test)) #bf_tda_kde_5.50.5_rf.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_rf.n5_test))
##Non-TDA-Assisted
svmGrid<-expand.grid(sigma = c(0.1, 1, 10), C = (1:5*0.25))
#Support Vector Machine-Radial Basis
dryBeanSvmFit <- train(as.factor(Class) ~ ., data = Dry_Bean_DatasetTrain,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
dryBeanSvmFit
## Support Vector Machines with Radial Basis Function Kernel
##
## 9531 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6353, 6355, 6354
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9268711 0.9115196
## 0.1 0.50 0.9275003 0.9122810
## 0.1 0.75 0.9290742 0.9141951
## 0.1 1.00 0.9291794 0.9143259
## 0.1 1.25 0.9292842 0.9144525
## 1.0 0.25 0.8977036 0.8761324
## 1.0 0.50 0.9112381 0.8925862
## 1.0 0.75 0.9159600 0.8983123
## 1.0 1.00 0.9178480 0.9006105
## 1.0 1.25 0.9181629 0.9009961
## 10.0 0.25 0.3487567 0.1295442
## 10.0 0.50 0.4499008 0.2742981
## 10.0 0.75 0.5237639 0.3786787
## 10.0 1.00 0.6038176 0.4900585
## 10.0 1.25 0.6312022 0.5277824
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
dryBeanSvmFit$resample
## Accuracy Kappa Resample
## 1 0.9238515 0.9078548 Fold1
## 2 0.9320113 0.9177619 Fold3
## 3 0.9319899 0.9177409 Fold2
db_svm_fit_re<-dryBeanSvmFit$resample[1]
summary(dryBeanSvmFit)
## Length Class Mode
## 1 ksvm S4
#vip(dryBeanSvmFit, 25) + ggtitle("non-TDA-Assited Svm")
# Predict outcome using model from training data based on testing data
predictions <- predict(dryBeanSvmFit, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_svm_cf<-confusionMatrix(data=predictions, as.factor(Dry_Bean_DatasetTest$Class))
db_svm_cf
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 366 0 12 0 0 4 3
## BOMBAY 1 156 0 0 0 0 0
## CALI 15 0 466 0 8 0 1
## DERMASON 0 0 0 971 6 10 80
## HOROZ 3 0 4 1 556 0 12
## SEKER 2 0 1 20 0 576 3
## SIRA 9 0 6 71 8 18 691
##
## Overall Statistics
##
## Accuracy : 0.927
## 95% CI : (0.9185, 0.9348)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9117
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.92424 1.00000 0.9530 0.9135
## Specificity 0.99484 0.99975 0.9933 0.9682
## Pos Pred Value 0.95065 0.99363 0.9510 0.9100
## Neg Pred Value 0.99188 1.00000 0.9936 0.9695
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08971 0.03824 0.1142 0.2380
## Detection Prevalence 0.09436 0.03848 0.1201 0.2615
## Balanced Accuracy 0.95954 0.99987 0.9731 0.9408
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9619 0.9474 0.8747
## Specificity 0.9943 0.9925 0.9660
## Pos Pred Value 0.9653 0.9568 0.8605
## Neg Pred Value 0.9937 0.9908 0.9698
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1363 0.1412 0.1694
## Detection Prevalence 0.1412 0.1475 0.1968
## Balanced Accuracy 0.9781 0.9699 0.9203
db_svm_cf$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9269608 0.9116543 0.9185443 0.9347595 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_svm_cf_ov_acc<-db_svm_cf$overall[1]
db_svm_cf$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9242424 0.9948426 0.9506494 0.9918809 0.9506494
## Class: BOMBAY 1.0000000 0.9997452 0.9936306 1.0000000 0.9936306
## Class: CALI 0.9529652 0.9933166 0.9510204 0.9935933 0.9510204
## Class: DERMASON 0.9134525 0.9681803 0.9100281 0.9694656 0.9100281
## Class: HOROZ 0.9619377 0.9942890 0.9652778 0.9937215 0.9652778
## Class: SEKER 0.9473684 0.9925115 0.9568106 0.9907993 0.9568106
## Class: SIRA 0.8746835 0.9659574 0.8605230 0.9697894 0.8605230
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9242424 0.9372599 0.09705882 0.08970588
## Class: BOMBAY 1.0000000 0.9968051 0.03823529 0.03823529
## Class: CALI 0.9529652 0.9519918 0.11985294 0.11421569
## Class: DERMASON 0.9134525 0.9117371 0.26053922 0.23799020
## Class: HOROZ 0.9619377 0.9636049 0.14166667 0.13627451
## Class: SEKER 0.9473684 0.9520661 0.14901961 0.14117647
## Class: SIRA 0.8746835 0.8675455 0.19362745 0.16936275
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09436275 0.9595425
## Class: BOMBAY 0.03848039 0.9998726
## Class: CALI 0.12009804 0.9731409
## Class: DERMASON 0.26151961 0.9408164
## Class: HOROZ 0.14117647 0.9781133
## Class: SEKER 0.14754902 0.9699400
## Class: SIRA 0.19681373 0.9203205
db_svm_cf_pr_rec_f1<-db_svm_cf$byClass[5:7]
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_PC_5.50.5_n1_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_PC_5.50.5_n1_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 7839 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5225, 5227, 5226
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9113419 0.8627017920
## 0.1 0.50 0.9135106 0.8661255376
## 0.1 0.75 0.9142758 0.8673248524
## 0.1 1.00 0.9146582 0.8679626002
## 0.1 1.25 0.9154233 0.8691065270
## 1.0 0.25 0.8748554 0.8028396839
## 1.0 0.50 0.8877401 0.8241868748
## 1.0 0.75 0.8957772 0.8372588298
## 1.0 1.00 0.8974359 0.8401116413
## 1.0 1.25 0.8987114 0.8422065390
## 10.0 0.25 0.4524811 0.0008332259
## 10.0 0.50 0.4730200 0.0452479889
## 10.0 0.75 0.5098869 0.1233115931
## 10.0 1.00 0.5533870 0.2136329066
## 10.0 1.25 0.5865545 0.2813179245
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_PC_5.50.5_n1_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9123948 0.8643032 Fold1
## 2 0.9184845 0.8740524 Fold3
## 3 0.9153905 0.8689640 Fold2
db_tda_pc_5.50.5_n1_svm_fit_re<-DryBean_TDA_PC_5.50.5_n1_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_PC_5.50.5_n1_SvmFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n1_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_PC_5.50.5_n1_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n1_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n1_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n1_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 53 0 0 0 0 2 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 277 156 455 996 533 8 101
## HOROZ 0 0 0 0 0 0 0
## SEKER 54 0 29 12 0 583 6
## SIRA 12 0 5 55 45 15 683
##
## Overall Statistics
##
## Accuracy : 0.5674
## 95% CI : (0.552, 0.5827)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4409
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.13384 0.00000 0.0000 0.9370
## Specificity 0.99946 1.00000 1.0000 0.4929
## Pos Pred Value 0.96364 NaN NaN 0.3943
## Neg Pred Value 0.91478 0.96176 0.8801 0.9569
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.01299 0.00000 0.0000 0.2441
## Detection Prevalence 0.01348 0.00000 0.0000 0.6191
## Balanced Accuracy 0.56665 0.50000 0.5000 0.7149
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.9589 0.8646
## Specificity 1.0000 0.9709 0.9599
## Pos Pred Value NaN 0.8523 0.8380
## Neg Pred Value 0.8583 0.9926 0.9672
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1429 0.1674
## Detection Prevalence 0.0000 0.1676 0.1998
## Balanced Accuracy 0.5000 0.9649 0.9122
db_tda_pc_5.50.5_n1_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 53 0 0 0 0 2 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 277 156 455 996 533 8 101
## HOROZ 0 0 0 0 0 0 0
## SEKER 54 0 29 12 0 583 6
## SIRA 12 0 5 55 45 15 683
##
## Overall Statistics
##
## Accuracy : 0.5674
## 95% CI : (0.552, 0.5827)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4409
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.13384 0.00000 0.0000 0.9370
## Specificity 0.99946 1.00000 1.0000 0.4929
## Pos Pred Value 0.96364 NaN NaN 0.3943
## Neg Pred Value 0.91478 0.96176 0.8801 0.9569
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.01299 0.00000 0.0000 0.2441
## Detection Prevalence 0.01348 0.00000 0.0000 0.6191
## Balanced Accuracy 0.56665 0.50000 0.5000 0.7149
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.9589 0.8646
## Specificity 1.0000 0.9709 0.9599
## Pos Pred Value NaN 0.8523 0.8380
## Neg Pred Value 0.8583 0.9926 0.9672
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1429 0.1674
## Detection Prevalence 0.0000 0.1676 0.1998
## Balanced Accuracy 0.5000 0.9649 0.9122
db_tda_pc_5.50.5_n1_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.5674020 0.4408902 0.5520335 0.5826738 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n1_db_svm_cf0_ov_acc<-db_tda_pc_5.50.5_n1_db_svm_cf0$overall[1]
db_tda_pc_5.50.5_n1_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.1338384 0.9994571 0.9636364 0.9147826 0.9636364
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.9369708 0.4928737 0.3942993 0.9568855 0.3942993
## Class: HOROZ 0.0000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.9588816 0.9709101 0.8523392 0.9926384 0.8523392
## Class: SIRA 0.8645570 0.9598784 0.8380368 0.9672282 0.8380368
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.1338384 0.2350333 0.09705882 0.0129902
## Class: BOMBAY 0.0000000 NA 0.03823529 0.0000000
## Class: CALI 0.0000000 NA 0.11985294 0.0000000
## Class: DERMASON 0.9369708 0.5550293 0.26053922 0.2441176
## Class: HOROZ 0.0000000 NA 0.14166667 0.0000000
## Class: SEKER 0.9588816 0.9024768 0.14901961 0.1428922
## Class: SIRA 0.8645570 0.8510903 0.19362745 0.1674020
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.01348039 0.5666477
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.00000000 0.5000000
## Class: DERMASON 0.61911765 0.7149223
## Class: HOROZ 0.00000000 0.5000000
## Class: SEKER 0.16764706 0.9648959
## Class: SIRA 0.19975490 0.9122177
db_tda_pc_5.50.5_n1_db_svm_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n1_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_svm_n1_3_fold<-(db_svm_fit_re - db_tda_pc_5.50.5_n1_svm_fit_re)
diff_drybean_tda_pca_5.50.5_svm_n1_3_fold
## Accuracy
## 1 0.01145668
## 2 0.01352683
## 3 0.01659942
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n1_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n1_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1681667
##
## $winRight
## [1] 0.8318333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n1_3_fold
## $left
## [1] 0.002593085
##
## $rope
## [1] 0.07471691
##
## $right
## [1] 0.92269
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold))
#bf_tda_pca_5.50.5_rf.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_svm_n1_3_fold)
## t = 9.2781, df = 2, p-value = 0.01142
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.007433021 0.020288934
## sample estimates:
## mean of x
## 0.01386098
### Test set diff
diff_drybean_tda_pca_5.50.5_svm.n1_test<-(db_svm_cf_ov_acc - db_tda_pc_5.50.5_n1_db_svm_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_svm.n1_test
## Accuracy
## 0.3595588
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n1_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n1_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n1_test$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n1_test$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n1_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1627333
##
## $winRight
## [1] 0.8372667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n1_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm.n1_test)))
#BayesFactor
#bf_tda_pca_5.50.5_svm.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm.n1_test)) #bf_tda_pca_5.50.5_svm.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n1_test))
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node2
DryBean_TDA_PC_5.50.5_n2_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_PC_5.50.5_n2_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 9515 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6344, 6343, 6343
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.6442739 0.59036543
## 0.1 0.50 0.6446941 0.59073778
## 0.1 0.75 0.6440636 0.58993080
## 0.1 1.00 0.6449044 0.59090851
## 0.1 1.25 0.6450095 0.59099313
## 1.0 0.25 0.6322921 0.57402305
## 1.0 0.50 0.6367059 0.57984296
## 1.0 0.75 0.6394384 0.58353174
## 1.0 1.00 0.6410151 0.58565467
## 1.0 1.25 0.6408050 0.58537991
## 10.0 0.25 0.2299623 0.03489697
## 10.0 0.50 0.2921832 0.12100639
## 10.0 0.75 0.3241348 0.16619282
## 10.0 1.00 0.3518817 0.20539323
## 10.0 1.25 0.3689081 0.22948435
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_PC_5.50.5_n2_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9101230 0.886555766 Fold1
## 2 0.1210593 0.007868807 Fold3
## 3 0.9038462 0.878554819 Fold2
db_tda_pc_5.50.5_n2_svm_fit_re<-DryBean_TDA_PC_5.50.5_n2_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n2_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_PC_5.50.5_n2_SvmFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n2_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_PC_5.50.5_n2_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n2_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n2_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n2_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 370 1 22 0 3 8 3
## BOMBAY 0 0 0 0 0 0 0
## CALI 12 0 451 0 5 0 1
## DERMASON 0 0 0 986 6 17 84
## HOROZ 3 155 10 1 556 2 8
## SEKER 3 0 1 16 0 564 3
## SIRA 8 0 5 60 8 17 691
##
## Overall Statistics
##
## Accuracy : 0.8868
## 95% CI : (0.8766, 0.8963)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8623
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.93434 0.00000 0.9223 0.9276
## Specificity 0.98996 1.00000 0.9950 0.9645
## Pos Pred Value 0.90909 NaN 0.9616 0.9021
## Neg Pred Value 0.99292 0.96176 0.9895 0.9742
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09069 0.00000 0.1105 0.2417
## Detection Prevalence 0.09975 0.00000 0.1150 0.2679
## Balanced Accuracy 0.96215 0.50000 0.9586 0.9460
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9619 0.9276 0.8747
## Specificity 0.9489 0.9934 0.9702
## Pos Pred Value 0.7565 0.9608 0.8758
## Neg Pred Value 0.9934 0.9874 0.9699
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1363 0.1382 0.1694
## Detection Prevalence 0.1801 0.1439 0.1934
## Balanced Accuracy 0.9554 0.9605 0.9224
db_tda_pc_5.50.5_n2_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 370 1 22 0 3 8 3
## BOMBAY 0 0 0 0 0 0 0
## CALI 12 0 451 0 5 0 1
## DERMASON 0 0 0 986 6 17 84
## HOROZ 3 155 10 1 556 2 8
## SEKER 3 0 1 16 0 564 3
## SIRA 8 0 5 60 8 17 691
##
## Overall Statistics
##
## Accuracy : 0.8868
## 95% CI : (0.8766, 0.8963)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8623
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.93434 0.00000 0.9223 0.9276
## Specificity 0.98996 1.00000 0.9950 0.9645
## Pos Pred Value 0.90909 NaN 0.9616 0.9021
## Neg Pred Value 0.99292 0.96176 0.9895 0.9742
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.09069 0.00000 0.1105 0.2417
## Detection Prevalence 0.09975 0.00000 0.1150 0.2679
## Balanced Accuracy 0.96215 0.50000 0.9586 0.9460
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9619 0.9276 0.8747
## Specificity 0.9489 0.9934 0.9702
## Pos Pred Value 0.7565 0.9608 0.8758
## Neg Pred Value 0.9934 0.9874 0.9699
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1363 0.1382 0.1694
## Detection Prevalence 0.1801 0.1439 0.1934
## Balanced Accuracy 0.9554 0.9605 0.9224
db_tda_pc_5.50.5_n2_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8867647 0.8623010 0.8766407 0.8963310 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n2_db_svm_cf0_ov_acc<-db_tda_pc_5.50.5_n2_db_svm_cf0$overall[1]
db_tda_pc_5.50.5_n2_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9343434 0.9899566 0.9090909 0.9929213 0.9090909
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.9222904 0.9949875 0.9616205 0.9894766 0.9616205
## Class: DERMASON 0.9275635 0.9645343 0.9021043 0.9742216 0.9021043
## Class: HOROZ 0.9619377 0.9488864 0.7564626 0.9934230 0.7564626
## Class: SEKER 0.9276316 0.9933756 0.9608177 0.9874034 0.9608177
## Class: SIRA 0.8746835 0.9702128 0.8757921 0.9699180 0.8757921
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9343434 0.9215442 0.09705882 0.09068627
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.9222904 0.9415449 0.11985294 0.11053922
## Class: DERMASON 0.9275635 0.9146568 0.26053922 0.24166667
## Class: HOROZ 0.9619377 0.8469155 0.14166667 0.13627451
## Class: SEKER 0.9276316 0.9439331 0.14901961 0.13823529
## Class: SIRA 0.8746835 0.8752375 0.19362745 0.16936275
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0997549 0.9621500
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.1149510 0.9586389
## Class: DERMASON 0.2678922 0.9460489
## Class: HOROZ 0.1801471 0.9554120
## Class: SEKER 0.1438725 0.9605036
## Class: SIRA 0.1933824 0.9224482
db_tda_pc_5.50.5_n2_db_svm_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n2_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_svm_n2_3_fold<-(db_svm_fit_re - db_tda_pc_5.50.5_n2_svm_fit_re)
diff_drybean_tda_pca_5.50.5_svm_n2_3_fold
## Accuracy
## 1 0.01372849
## 2 0.81095206
## 3 0.02814377
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n2_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n2_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.03776667
##
## $winRight
## [1] 0.9622333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n2_3_fold
## $left
## [1] 0.2176536
##
## $rope
## [1] 0.01350065
##
## $right
## [1] 0.7688458
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold))
#bf_tda_pca_5.50.5_rf.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_svm_n2_3_fold)
## t = 1.0794, df = 2, p-value = 0.3933
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.8489214 1.4174710
## sample estimates:
## mean of x
## 0.2842748
### Test set diff
diff_drybean_tda_pca_5.50.5_svm.n2_test<-(db_svm_cf_ov_acc - db_tda_pc_5.50.5_n2_db_svm_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_svm.n2_test
## Accuracy
## 0.04019608
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n2_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n2_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n2_test$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n2_test$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n2_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1584
##
## $winRight
## [1] 0.8416
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n2_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm.n2_test)))
#BayesFactor
#bf_tda_pca_5.50.5_svm.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm.n2_test)) #bf_tda_pca_5.50.5_svm.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n2_test))
##Node3
DryBean_TDA_PC_5.50.5_n3_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_PC_5.50.5_n3_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 5355 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 3569, 3572, 3569
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9443508 0.922342078
## 0.1 0.50 0.9460328 0.924707408
## 0.1 0.75 0.9462194 0.924997413
## 0.1 1.00 0.9467790 0.925787231
## 0.1 1.25 0.9473395 0.926572092
## 1.0 0.25 0.9116683 0.875162547
## 1.0 0.50 0.9254905 0.895305238
## 1.0 0.75 0.9299716 0.901796732
## 1.0 1.00 0.9329603 0.906046503
## 1.0 1.25 0.9329600 0.906138271
## 10.0 0.25 0.3555556 0.000000000
## 10.0 0.50 0.3598510 0.007010944
## 10.0 0.75 0.4089632 0.087523580
## 10.0 1.00 0.5069990 0.246940012
## 10.0 1.25 0.5353810 0.293893563
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_PC_5.50.5_n3_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9524076 0.9337115 Fold1
## 2 0.9412094 0.9179588 Fold3
## 3 0.9484016 0.9280459 Fold2
db_tda_pc_5.50.5_n3_svm_fit_re<-DryBean_TDA_PC_5.50.5_n3_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n3_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_PC_5.50.5_n3_SvmFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n3_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_PC_5.50.5_n3_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n3_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n3_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n3_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 372 0 14 44 0 59 143
## BOMBAY 0 83 0 0 0 0 0
## CALI 17 0 467 0 8 0 1
## DERMASON 0 0 0 1 0 0 0
## HOROZ 4 73 4 987 561 546 111
## SEKER 0 0 0 0 0 0 0
## SIRA 3 0 4 31 9 3 535
##
## Overall Statistics
##
## Accuracy : 0.4949
## 95% CI : (0.4794, 0.5103)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4143
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.93939 0.53205 0.9550 0.0009407
## Specificity 0.92942 1.00000 0.9928 1.0000000
## Pos Pred Value 0.58861 1.00000 0.9473 1.0000000
## Neg Pred Value 0.99304 0.98174 0.9939 0.7396421
## Prevalence 0.09706 0.03824 0.1199 0.2605392
## Detection Rate 0.09118 0.02034 0.1145 0.0002451
## Detection Prevalence 0.15490 0.02034 0.1208 0.0002451
## Balanced Accuracy 0.93441 0.76603 0.9739 0.5004704
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9706 0.000 0.6772
## Specificity 0.5074 1.000 0.9848
## Pos Pred Value 0.2454 NaN 0.9145
## Neg Pred Value 0.9905 0.851 0.9270
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1375 0.000 0.1311
## Detection Prevalence 0.5603 0.000 0.1434
## Balanced Accuracy 0.7390 0.500 0.8310
db_tda_pc_5.50.5_n3_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 372 0 14 44 0 59 143
## BOMBAY 0 83 0 0 0 0 0
## CALI 17 0 467 0 8 0 1
## DERMASON 0 0 0 1 0 0 0
## HOROZ 4 73 4 987 561 546 111
## SEKER 0 0 0 0 0 0 0
## SIRA 3 0 4 31 9 3 535
##
## Overall Statistics
##
## Accuracy : 0.4949
## 95% CI : (0.4794, 0.5103)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4143
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.93939 0.53205 0.9550 0.0009407
## Specificity 0.92942 1.00000 0.9928 1.0000000
## Pos Pred Value 0.58861 1.00000 0.9473 1.0000000
## Neg Pred Value 0.99304 0.98174 0.9939 0.7396421
## Prevalence 0.09706 0.03824 0.1199 0.2605392
## Detection Rate 0.09118 0.02034 0.1145 0.0002451
## Detection Prevalence 0.15490 0.02034 0.1208 0.0002451
## Balanced Accuracy 0.93441 0.76603 0.9739 0.5004704
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9706 0.000 0.6772
## Specificity 0.5074 1.000 0.9848
## Pos Pred Value 0.2454 NaN 0.9145
## Neg Pred Value 0.9905 0.851 0.9270
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1375 0.000 0.1311
## Detection Prevalence 0.5603 0.000 0.1434
## Balanced Accuracy 0.7390 0.500 0.8310
db_tda_pc_5.50.5_n3_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.948529e-01 4.143248e-01 4.793973e-01 5.103159e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 7.591924e-224 NaN
db_tda_pc_5.50.5_n3_db_svm_cf0_ov_acc<-db_tda_pc_5.50.5_n3_db_svm_cf0$overall[1]
db_tda_pc_5.50.5_n3_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0.9393939394 0.9294245 0.5886076 0.9930394
## Class: BOMBAY 0.5320512821 1.0000000 1.0000000 0.9817363
## Class: CALI 0.9550102249 0.9927597 0.9472617 0.9938667
## Class: DERMASON 0.0009407338 1.0000000 1.0000000 0.7396421
## Class: HOROZ 0.9705882353 0.5074243 0.2454068 0.9905240
## Class: SEKER 0.0000000000 1.0000000 NaN 0.8509804
## Class: SIRA 0.6772151899 0.9848024 0.9145299 0.9270386
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.5886076 0.9393939394 0.723735409 0.09705882 0.091176471
## Class: BOMBAY 1.0000000 0.5320512821 0.694560669 0.03823529 0.020343137
## Class: CALI 0.9472617 0.9550102249 0.951120163 0.11985294 0.114460784
## Class: DERMASON 1.0000000 0.0009407338 0.001879699 0.26053922 0.000245098
## Class: HOROZ 0.2454068 0.9705882353 0.391759777 0.14166667 0.137500000
## Class: SEKER NA 0.0000000000 NA 0.14901961 0.000000000
## Class: SIRA 0.9145299 0.6772151899 0.778181818 0.19362745 0.131127451
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.154901961 0.9344092
## Class: BOMBAY 0.020343137 0.7660256
## Class: CALI 0.120833333 0.9738850
## Class: DERMASON 0.000245098 0.5004704
## Class: HOROZ 0.560294118 0.7390063
## Class: SEKER 0.000000000 0.5000000
## Class: SIRA 0.143382353 0.8310088
db_tda_pc_5.50.5_n3_db_svm_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n3_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_svm_n3_3_fold<-(db_svm_fit_re - db_tda_pc_5.50.5_n3_svm_fit_re)
diff_drybean_tda_pca_5.50.5_svm_n3_3_fold
## Accuracy
## 1 -0.028556136
## 2 -0.009198075
## 3 -0.016411646
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold
## $probLeft
## [1] 0.5
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n3_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n3_3_fold
## $winLeft
## [1] 0.7847
##
## $winRope
## [1] 0.2153
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n3_3_fold
## $left
## [1] 0.8288962
##
## $rope
## [1] 0.1460922
##
## $right
## [1] 0.02501158
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold))
#bf_tda_pca_5.50.5_rf.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_svm_n3_3_fold)
## t = -3.1966, df = 2, p-value = 0.0855
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.042357948 0.006247377
## sample estimates:
## mean of x
## -0.01805529
### Test set diff
diff_drybean_tda_pca_5.50.5_svm.n3_test<-(db_svm_cf_ov_acc - db_tda_pc_5.50.5_n3_db_svm_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_svm.n3_test
## Accuracy
## 0.4321078
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n3_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n3_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n3_test$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n3_test$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n3_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1607
##
## $winRight
## [1] 0.8393
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n3_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm.n3_test)))
#BayesFactor
#bf_tda_pca_5.50.5_svm.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm.n3_test)) #bf_tda_pca_5.50.5_svm.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n3_test))
##Node4
DryBean_TDA_PC_5.50.5_n4_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_PC_5.50.5_n4_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1060, 1059, 1061
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9779743 0.96780604
## 0.1 0.50 0.9792310 0.96966610
## 0.1 0.75 0.9792298 0.96967461
## 0.1 1.00 0.9792298 0.96967461
## 0.1 1.25 0.9811190 0.97245730
## 1.0 0.25 0.9383634 0.90870262
## 1.0 0.50 0.9559688 0.93561096
## 1.0 0.75 0.9609931 0.94301438
## 1.0 1.00 0.9653957 0.94943421
## 1.0 1.25 0.9653980 0.94945911
## 10.0 0.25 0.4201256 0.00000000
## 10.0 0.50 0.4213846 0.00253303
## 10.0 0.75 0.4446599 0.04877826
## 10.0 1.00 0.4905684 0.13663762
## 10.0 1.25 0.5257862 0.20388946
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_PC_5.50.5_n4_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9867925 0.9808476 Fold1
## 2 0.9678639 0.9530419 Fold3
## 3 0.9887006 0.9834825 Fold2
db_tda_pc_5.50.5_n4_svm_fit_re<-DryBean_TDA_PC_5.50.5_n4_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n4_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_PC_5.50.5_n4_SvmFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n4_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_PC_5.50.5_n4_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n4_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n4_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n4_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 252 0 6 0 0 0 1
## BOMBAY 5 156 0 0 0 0 0
## CALI 31 0 462 0 22 0 10
## DERMASON 0 0 0 0 0 0 0
## HOROZ 108 0 21 1063 556 608 779
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3495
## 95% CI : (0.3349, 0.3644)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2506
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.63636 1.00000 0.9448 0.0000
## Specificity 0.99810 0.99873 0.9825 1.0000
## Pos Pred Value 0.97297 0.96894 0.8800 NaN
## Neg Pred Value 0.96231 1.00000 0.9924 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.06176 0.03824 0.1132 0.0000
## Detection Prevalence 0.06348 0.03946 0.1287 0.0000
## Balanced Accuracy 0.81723 0.99936 0.9636 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9619 0.000 0.0000
## Specificity 0.2636 1.000 1.0000
## Pos Pred Value 0.1774 NaN NaN
## Neg Pred Value 0.9767 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1363 0.000 0.0000
## Detection Prevalence 0.7684 0.000 0.0000
## Balanced Accuracy 0.6128 0.500 0.5000
db_tda_pc_5.50.5_n4_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 252 0 6 0 0 0 1
## BOMBAY 5 156 0 0 0 0 0
## CALI 31 0 462 0 22 0 10
## DERMASON 0 0 0 0 0 0 0
## HOROZ 108 0 21 1063 556 608 779
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3495
## 95% CI : (0.3349, 0.3644)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2506
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.63636 1.00000 0.9448 0.0000
## Specificity 0.99810 0.99873 0.9825 1.0000
## Pos Pred Value 0.97297 0.96894 0.8800 NaN
## Neg Pred Value 0.96231 1.00000 0.9924 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.06176 0.03824 0.1132 0.0000
## Detection Prevalence 0.06348 0.03946 0.1287 0.0000
## Balanced Accuracy 0.81723 0.99936 0.9636 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9619 0.000 0.0000
## Specificity 0.2636 1.000 1.0000
## Pos Pred Value 0.1774 NaN NaN
## Neg Pred Value 0.9767 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1363 0.000 0.0000
## Detection Prevalence 0.7684 0.000 0.0000
## Balanced Accuracy 0.6128 0.500 0.5000
db_tda_pc_5.50.5_n4_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.495098e-01 2.506335e-01 3.348685e-01 3.643669e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 2.511382e-36 NaN
db_tda_pc_5.50.5_n4_db_svm_cf0_ov_acc<-db_tda_pc_5.50.5_n4_db_svm_cf0$overall[1]
db_tda_pc_5.50.5_n4_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.6363636 0.9980999 0.9729730 0.9623135 0.9729730
## Class: BOMBAY 1.0000000 0.9987258 0.9689441 1.0000000 0.9689441
## Class: CALI 0.9447853 0.9824561 0.8800000 0.9924051 0.8800000
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9619377 0.2635637 0.1773525 0.9767196 0.1773525
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.6363636 0.7694656 0.09705882 0.06176471
## Class: BOMBAY 1.0000000 0.9842271 0.03823529 0.03823529
## Class: CALI 0.9447853 0.9112426 0.11985294 0.11323529
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9619377 0.2994883 0.14166667 0.13627451
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.06348039 0.8172318
## Class: BOMBAY 0.03946078 0.9993629
## Class: CALI 0.12867647 0.9636207
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.76838235 0.6127507
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
db_tda_pc_5.50.5_n4_db_svm_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n4_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_svm_n4_3_fold<-(db_svm_fit_re - db_tda_pc_5.50.5_n4_svm_fit_re)
diff_drybean_tda_pca_5.50.5_svm_n4_3_fold
## Accuracy
## 1 -0.06294097
## 2 -0.03585256
## 3 -0.05671064
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n4_3_fold
## $winLeft
## [1] 0.9916333
##
## $winRope
## [1] 0.008366667
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n4_3_fold
## $left
## [1] 0.9762494
##
## $rope
## [1] 0.01244765
##
## $right
## [1] 0.01130295
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold))
#bf_tda_pca_5.50.5_rf.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_svm_n4_3_fold)
## t = -6.3283, df = 2, p-value = 0.02407
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.08707766 -0.01659179
## sample estimates:
## mean of x
## -0.05183473
### Test set diff
diff_drybean_tda_pca_5.50.5_svm.n4_test<-(db_svm_cf_ov_acc - db_tda_pc_5.50.5_n4_db_svm_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_svm.n4_test
## Accuracy
## 0.577451
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n4_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n4_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n4_test$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n4_test$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n4_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1543667
##
## $winRight
## [1] 0.8456333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n4_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm.n4_test)))
#BayesFactor
#bf_tda_pca_5.50.5_svm.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm.n4_test)) #bf_tda_pca_5.50.5_svm.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n4_test))
##Node5
DryBean_TDA_PC_5.50.5_n5_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
## Warning: model fit failed for Fold1: sigma= 0.1, C=0.25 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 1.0, C=0.25 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma=10.0, C=0.25 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 0.1, C=0.50 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 1.0, C=0.50 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma=10.0, C=0.50 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 0.1, C=0.75 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 1.0, C=0.75 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma=10.0, C=0.75 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 0.1, C=1.00 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 1.0, C=1.00 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma=10.0, C=1.00 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 0.1, C=1.25 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma= 1.0, C=1.25 Error in indexes[[j]] : subscript out of bounds
## Warning: model fit failed for Fold1: sigma=10.0, C=1.25 Error in indexes[[j]] : subscript out of bounds
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
DryBean_TDA_PC_5.50.5_n5_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 417 samples
## 16 predictor
## 2 classes: 'BOMBAY', 'CALI'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 277, 278, 279
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 1 NaN
## 0.1 0.50 1 NaN
## 0.1 0.75 1 NaN
## 0.1 1.00 1 NaN
## 0.1 1.25 1 NaN
## 1.0 0.25 1 NaN
## 1.0 0.50 1 NaN
## 1.0 0.75 1 NaN
## 1.0 1.00 1 NaN
## 1.0 1.25 1 NaN
## 10.0 0.25 1 NaN
## 10.0 0.50 1 NaN
## 10.0 0.75 1 NaN
## 10.0 1.00 1 NaN
## 10.0 1.25 1 NaN
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 10 and C = 0.25.
DryBean_TDA_PC_5.50.5_n5_SvmFit0$resample
## Accuracy Kappa Resample
## 1 NA NA Fold1
## 2 1 NA Fold2
## 3 1 NA Fold3
db_tda_pc_5.50.5_n5_svm_fit_re<-DryBean_TDA_PC_5.50.5_n5_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n5_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_PC_5.50.5_n5_SvmFit0,25) + ggtitle("Adult_TDA_PCA_5.50.5_n5_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_PC_5.50.5_n5_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n5_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n5_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n5_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.03823529 0.00000000 0.03256139 0.04458199 0.26053922
## AccuracyPValue McnemarPValue
## 1.00000000 NaN
db_tda_pc_5.50.5_n5_db_svm_cf0_ov_acc<-db_tda_pc_5.50.5_n5_db_svm_cf0$overall[1]
db_tda_pc_5.50.5_n5_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0 1 NaN 0.9029412
## Class: BOMBAY 1 0 0.03823529 NaN
## Class: CALI 0 1 NaN 0.8801471
## Class: DERMASON 0 1 NaN 0.7394608
## Class: HOROZ 0 1 NaN 0.8583333
## Class: SEKER 0 1 NaN 0.8509804
## Class: SIRA 0 1 NaN 0.8063725
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA NA 0 NA 0.09705882 0.00000000
## Class: BOMBAY 0.03823529 1 0.07365439 0.03823529 0.03823529
## Class: CALI NA 0 NA 0.11985294 0.00000000
## Class: DERMASON NA 0 NA 0.26053922 0.00000000
## Class: HOROZ NA 0 NA 0.14166667 0.00000000
## Class: SEKER NA 0 NA 0.14901961 0.00000000
## Class: SIRA NA 0 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0 0.5
## Class: BOMBAY 1 0.5
## Class: CALI 0 0.5
## Class: DERMASON 0 0.5
## Class: HOROZ 0 0.5
## Class: SEKER 0 0.5
## Class: SIRA 0 0.5
db_tda_pc_5.50.5_n5_db_svm_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n5_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_svm_n5_3_fold<-(db_svm_fit_re - db_tda_pc_5.50.5_n5_svm_fit_re)
diff_drybean_tda_pca_5.50.5_svm_n5_3_fold
## Accuracy
## 1 NA
## 2 -0.06798867
## 3 -0.06801008
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold
## $probLeft
## [1] NA
##
## $probRope
## [1] NA
##
## $probRight
## [1] NA
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n5_3_fold_odds.left
## [1] NA
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_svm.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_svm.n5_3_fold
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n5_3_fold
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_rf.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold))
#bf_tda_pca_5.50.5_rf.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_svm_n5_3_fold)
## t = -6353, df = 1, p-value = 0.0001002
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.06813537 -0.06786337
## sample estimates:
## mean of x
## -0.06799937
### Test set diff
diff_drybean_tda_pca_5.50.5_svm.n5_test<-(db_svm_cf_ov_acc - db_tda_pc_5.50.5_n5_db_svm_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_svm.n5_test
## Accuracy
## 0.8887255
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n5_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_svm.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_svm.n5_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_svm.n5_test$probLeft/bst_dbf_db_tda_pca_5.50.5_svm.n5_test$probRight
bst_dbf_db_tda_pca_5.50.5_svm.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_svm.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n5_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_svm.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1587333
##
## $winRight
## [1] 0.8412667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_svm.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n5_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_svm.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_svm.n5_test)))
#BayesFactor
#bf_tda_pca_5.50.5_svm.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_svm.n5_test)) #bf_tda_pca_5.50.5_svm.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_svm.n5_test))
#With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_KDE_5.50.5_n1_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_KDE_5.50.5_n1_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 8473 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5650, 5648, 5648
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9523182 0.9430179
## 0.1 0.50 0.9532625 0.9441534
## 0.1 0.75 0.9534985 0.9444374
## 0.1 1.00 0.9542066 0.9452842
## 0.1 1.25 0.9550329 0.9462702
## 1.0 0.25 0.9261169 0.9113623
## 1.0 0.50 0.9378005 0.9255079
## 1.0 0.75 0.9437018 0.9326311
## 1.0 1.00 0.9457080 0.9350491
## 1.0 1.25 0.9460622 0.9354778
## 10.0 0.25 0.3028453 0.1127524
## 10.0 0.50 0.4405774 0.2959555
## 10.0 0.75 0.5591906 0.4501030
## 10.0 1.00 0.6531363 0.5705731
## 10.0 1.25 0.6768589 0.6009978
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_KDE_5.50.5_n1_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9518243 0.9424457 Fold1
## 2 0.9582301 0.9500843 Fold3
## 3 0.9550442 0.9462807 Fold2
ad_tda_kde_5.50.5_n1_svm_fit_re<-DryBean_TDA_KDE_5.50.5_n1_SvmFit0 $resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_KDE_5.50.5_n1_SvmFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n1_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n1_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n1_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n1_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
ad_tda_kde_5.50.5_n1_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 10 0 1 3 3
## BOMBAY 0 156 0 0 0 0 0
## CALI 17 0 469 0 7 0 1
## DERMASON 0 0 0 861 4 7 28
## HOROZ 3 0 4 1 558 0 9
## SEKER 3 0 1 19 0 577 4
## SIRA 9 0 5 182 8 21 745
##
## Overall Statistics
##
## Accuracy : 0.9142
## 95% CI : (0.9052, 0.9226)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8966
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9591 0.8100
## Specificity 0.99539 1.00000 0.9930 0.9871
## Pos Pred Value 0.95538 1.00000 0.9494 0.9567
## Neg Pred Value 0.99135 1.00000 0.9944 0.9365
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1150 0.2110
## Detection Prevalence 0.09338 0.03824 0.1211 0.2206
## Balanced Accuracy 0.95729 1.00000 0.9761 0.8985
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9654 0.9490 0.9430
## Specificity 0.9951 0.9922 0.9316
## Pos Pred Value 0.9704 0.9553 0.7680
## Neg Pred Value 0.9943 0.9911 0.9855
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1368 0.1414 0.1826
## Detection Prevalence 0.1409 0.1480 0.2377
## Balanced Accuracy 0.9803 0.9706 0.9373
ad_tda_kde_5.50.5_n1_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 10 0 1 3 3
## BOMBAY 0 156 0 0 0 0 0
## CALI 17 0 469 0 7 0 1
## DERMASON 0 0 0 861 4 7 28
## HOROZ 3 0 4 1 558 0 9
## SEKER 3 0 1 19 0 577 4
## SIRA 9 0 5 182 8 21 745
##
## Overall Statistics
##
## Accuracy : 0.9142
## 95% CI : (0.9052, 0.9226)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8966
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9591 0.8100
## Specificity 0.99539 1.00000 0.9930 0.9871
## Pos Pred Value 0.95538 1.00000 0.9494 0.9567
## Neg Pred Value 0.99135 1.00000 0.9944 0.9365
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1150 0.2110
## Detection Prevalence 0.09338 0.03824 0.1211 0.2206
## Balanced Accuracy 0.95729 1.00000 0.9761 0.8985
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9654 0.9490 0.9430
## Specificity 0.9951 0.9922 0.9316
## Pos Pred Value 0.9704 0.9553 0.7680
## Neg Pred Value 0.9943 0.9911 0.9855
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1368 0.1414 0.1826
## Detection Prevalence 0.1409 0.1480 0.2377
## Balanced Accuracy 0.9803 0.9706 0.9373
ad_tda_kde_5.50.5_n1_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9142157 0.8965745 0.9052005 0.9226324 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n1_db_svm_cf0_ov_acc<-ad_tda_kde_5.50.5_n1_db_svm_cf0$overall[1]
ad_tda_kde_5.50.5_n1_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.9953855 0.9553806 0.9913490 0.9553806
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9591002 0.9930382 0.9493927 0.9944228 0.9493927
## Class: DERMASON 0.8099718 0.9870733 0.9566667 0.9364780 0.9566667
## Class: HOROZ 0.9653979 0.9951456 0.9704348 0.9942939 0.9704348
## Class: SEKER 0.9490132 0.9922235 0.9552980 0.9910817 0.9552980
## Class: SIRA 0.9430380 0.9316109 0.7680412 0.9855305 0.7680412
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.9369369 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9591002 0.9542218 0.11985294 0.11495098
## Class: DERMASON 0.8099718 0.8772287 0.26053922 0.21102941
## Class: HOROZ 0.9653979 0.9679098 0.14166667 0.13676471
## Class: SEKER 0.9490132 0.9521452 0.14901961 0.14142157
## Class: SIRA 0.9430380 0.8465909 0.19362745 0.18259804
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09338235 0.9572887
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.12107843 0.9760692
## Class: DERMASON 0.22058824 0.8985225
## Class: HOROZ 0.14093137 0.9802718
## Class: SEKER 0.14803922 0.9706183
## Class: SIRA 0.23774510 0.9373245
ad_tda_kde_5.50.5_n1_db_svm_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n1_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_svm_n1_3_fold<-(db_svm_fit_re - ad_tda_kde_5.50.5_n1_svm_fit_re)
diff_drybean_tda_kde_5.50.5_svm_n1_3_fold
## Accuracy
## 1 -0.02797282
## 2 -0.02621876
## 3 -0.02305432
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n1_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n1_3_fold_odds.left<-bst_tda_kde_5.50.5_svm.n1_3_fold$probLeft/bst_tda_kde_5.50.5_svm.n1_3_fold$probRight
bst_tda_kde_5.50.5_svm.n1_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n1_3_fold
## $winLeft
## [1] 0.9913667
##
## $winRope
## [1] 0.008633333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n1_3_fold
## $left
## [1] 0.9945239
##
## $rope
## [1] 0.004399082
##
## $right
## [1] 0.00107699
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold))
#bf_tda_kde_5.50.5_svm.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_svm_n1_3_fold)
## t = -17.891, df = 2, p-value = 0.003109
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.03194090 -0.01955637
## sample estimates:
## mean of x
## -0.02574863
### Test set diff
diff_drybean_tda_kde_5.50.5_svm.n1_test<-(db_svm_cf_ov_acc-ad_tda_kde_5.50.5_n1_db_svm_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_svm.n1_test
## Accuracy
## 0.0127451
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n1_test),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n1_test_odds.left<-bst_tda_kde_5.50.5_svm.n1_test$probLeft/bst_tda_kde_5.50.5_svm.n1_test$probRight
bst_tda_kde_5.50.5_svm.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n1_test),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.4621333
##
## $winRight
## [1] 0.5378667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n1_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_svm.n1_test))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm.n1_test)) #bf_tda_kde_5.50.5_svm.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n1_test))
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node2
DryBean_TDA_KDE_5.50.5_n2_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_KDE_5.50.5_n2_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 7582 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5055, 5055, 5054
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9422319 0.9260614
## 0.1 0.50 0.9450013 0.9296485
## 0.1 0.75 0.9453971 0.9301714
## 0.1 1.00 0.9459246 0.9308563
## 0.1 1.25 0.9465842 0.9317000
## 1.0 0.25 0.9268006 0.9064155
## 1.0 0.50 0.9322083 0.9134002
## 1.0 0.75 0.9360331 0.9182953
## 1.0 1.00 0.9373520 0.9199733
## 1.0 1.25 0.9377475 0.9204803
## 10.0 0.25 0.5129282 0.3219701
## 10.0 0.50 0.6027430 0.4545282
## 10.0 0.75 0.6351886 0.5017353
## 10.0 1.00 0.6717231 0.5542406
## 10.0 1.25 0.6916382 0.5830410
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_KDE_5.50.5_n2_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9469727 0.9322224 Fold1
## 2 0.9450158 0.9296774 Fold3
## 3 0.9477641 0.9332001 Fold2
ad_tda_kde_5.50.5_n2_svm_fit_re<-DryBean_TDA_KDE_5.50.5_n2_SvmFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n2_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_KDE_5.50.5_n2_SvmFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n2_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n2_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n2_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n2_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n2_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 341 0 12 0 1 0 3
## BOMBAY 0 0 0 0 0 0 0
## CALI 26 0 461 0 21 0 1
## DERMASON 0 0 0 934 6 6 57
## HOROZ 16 156 8 1 543 0 8
## SEKER 3 0 1 18 0 581 3
## SIRA 10 0 7 110 7 21 718
##
## Overall Statistics
##
## Accuracy : 0.877
## 95% CI : (0.8665, 0.8869)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8506
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.86111 0.00000 0.9427 0.8786
## Specificity 0.99566 1.00000 0.9866 0.9771
## Pos Pred Value 0.95518 NaN 0.9057 0.9312
## Neg Pred Value 0.98523 0.96176 0.9922 0.9581
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08358 0.00000 0.1130 0.2289
## Detection Prevalence 0.08750 0.00000 0.1248 0.2458
## Balanced Accuracy 0.92838 0.50000 0.9647 0.9279
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9394 0.9556 0.9089
## Specificity 0.9460 0.9928 0.9529
## Pos Pred Value 0.7418 0.9587 0.8225
## Neg Pred Value 0.9895 0.9922 0.9775
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1331 0.1424 0.1760
## Detection Prevalence 0.1794 0.1485 0.2140
## Balanced Accuracy 0.9427 0.9742 0.9309
ad_tda_kde_5.50.5_n2_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 341 0 12 0 1 0 3
## BOMBAY 0 0 0 0 0 0 0
## CALI 26 0 461 0 21 0 1
## DERMASON 0 0 0 934 6 6 57
## HOROZ 16 156 8 1 543 0 8
## SEKER 3 0 1 18 0 581 3
## SIRA 10 0 7 110 7 21 718
##
## Overall Statistics
##
## Accuracy : 0.877
## 95% CI : (0.8665, 0.8869)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8506
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.86111 0.00000 0.9427 0.8786
## Specificity 0.99566 1.00000 0.9866 0.9771
## Pos Pred Value 0.95518 NaN 0.9057 0.9312
## Neg Pred Value 0.98523 0.96176 0.9922 0.9581
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08358 0.00000 0.1130 0.2289
## Detection Prevalence 0.08750 0.00000 0.1248 0.2458
## Balanced Accuracy 0.92838 0.50000 0.9647 0.9279
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9394 0.9556 0.9089
## Specificity 0.9460 0.9928 0.9529
## Pos Pred Value 0.7418 0.9587 0.8225
## Neg Pred Value 0.9895 0.9922 0.9775
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1331 0.1424 0.1760
## Detection Prevalence 0.1794 0.1485 0.2140
## Balanced Accuracy 0.9427 0.9742 0.9309
ad_tda_kde_5.50.5_n2_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8769608 0.8505944 0.8664883 0.8868901 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n2_db_svm_cf0_ov_acc<-ad_tda_kde_5.50.5_n2_db_svm_cf0$overall[1]
ad_tda_kde_5.50.5_n2_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.8611111 0.9956569 0.9551821 0.9852270 0.9551821
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.9427403 0.9866332 0.9056974 0.9921591 0.9056974
## Class: DERMASON 0.8786453 0.9771296 0.9312064 0.9580760 0.9312064
## Class: HOROZ 0.9394464 0.9460308 0.7418033 0.9895460 0.7418033
## Class: SEKER 0.9555921 0.9927995 0.9587459 0.9922280 0.9587459
## Class: SIRA 0.9088608 0.9528875 0.8224513 0.9775491 0.8224513
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.8611111 0.9057105 0.09705882 0.08357843
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.9427403 0.9238477 0.11985294 0.11299020
## Class: DERMASON 0.8786453 0.9041626 0.26053922 0.22892157
## Class: HOROZ 0.9394464 0.8290076 0.14166667 0.13308824
## Class: SEKER 0.9555921 0.9571664 0.14901961 0.14240196
## Class: SIRA 0.9088608 0.8634997 0.19362745 0.17598039
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0875000 0.9283840
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.1247549 0.9646868
## Class: DERMASON 0.2458333 0.9278875
## Class: HOROZ 0.1794118 0.9427386
## Class: SEKER 0.1485294 0.9741958
## Class: SIRA 0.2139706 0.9308741
ad_tda_kde_5.50.5_n2_db_svm_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n2_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_svm_n2_3_fold<-(db_svm_fit_re - ad_tda_kde_5.50.5_n2_svm_fit_re)
diff_drybean_tda_kde_5.50.5_svm_n2_3_fold
## Accuracy
## 1 -0.02312122
## 2 -0.01300449
## 3 -0.01577422
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n2_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n2_3_fold_odds.left<-bst_tda_kde_5.50.5_svm.n2_3_fold$probLeft/bst_tda_kde_5.50.5_svm.n2_3_fold$probRight
bst_tda_kde_5.50.5_svm.n2_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n2_3_fold
## $winLeft
## [1] 0.9056667
##
## $winRope
## [1] 0.09433333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n2_3_fold
## $left
## [1] 0.9143819
##
## $rope
## [1] 0.07766221
##
## $right
## [1] 0.007955856
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold))
#bf_tda_kde_5.50.5_svm.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_svm_n2_3_fold)
## t = -5.7314, df = 2, p-value = 0.02912
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.03028728 -0.00431267
## sample estimates:
## mean of x
## -0.01729998
### Test set diff
diff_drybean_tda_kde_5.50.5_svm.n2_test<-(db_svm_cf_ov_acc-ad_tda_kde_5.50.5_n2_db_svm_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_svm.n2_test
## Accuracy
## 0.05
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n2_test),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n2_test_odds.left<-bst_tda_kde_5.50.5_svm.n2_test$probLeft/bst_tda_kde_5.50.5_svm.n2_test$probRight
bst_tda_kde_5.50.5_svm.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n2_test),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1578333
##
## $winRight
## [1] 0.8421667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n2_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_svm.n2_test))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm.n2_test)) #bf_tda_kde_5.50.5_svm.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n2_test))
##Node3
DryBean_TDA_KDE_5.50.5_n3_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_KDE_5.50.5_n3_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 4149 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 2767, 2766, 2765
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.9113039 0.8656083559
## 0.1 0.50 0.9161243 0.8730287159
## 0.1 0.75 0.9163659 0.8734237721
## 0.1 1.00 0.9170891 0.8745612562
## 0.1 1.25 0.9178120 0.8756815811
## 1.0 0.25 0.8843126 0.8233130802
## 1.0 0.50 0.8939526 0.8383262657
## 1.0 0.75 0.8975674 0.8440411048
## 1.0 1.00 0.8990135 0.8464489877
## 1.0 1.25 0.8992547 0.8469257770
## 10.0 0.25 0.3856352 0.0004507734
## 10.0 0.50 0.4160038 0.0561045848
## 10.0 0.75 0.5049399 0.2087131185
## 10.0 1.00 0.6271389 0.4110371831
## 10.0 1.25 0.6560630 0.4588660759
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_KDE_5.50.5_n3_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.9225760 0.8828739 Fold1
## 2 0.9205202 0.8799162 Fold3
## 3 0.9103398 0.8642546 Fold2
ad_tda_kde_5.50.5_n3_svm_fit_re<-DryBean_TDA_KDE_5.50.5_n3_SvmFit0 $resample[1]
summary(DryBean_TDA_KDE_5.50.5_n3_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_KDE_5.50.5_n3_SvmFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n3_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n3_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n3_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n3_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n3_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 34 0 0 0 0 0 2
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 347 156 475 992 184 30 96
## HOROZ 3 0 9 0 369 0 5
## SEKER 1 0 0 17 0 553 2
## SIRA 11 0 5 54 25 25 685
##
## Overall Statistics
##
## Accuracy : 0.6453
## 95% CI : (0.6304, 0.66)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5459
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.085859 0.00000 0.0000 0.9332
## Specificity 0.999457 1.00000 1.0000 0.5731
## Pos Pred Value 0.944444 NaN NaN 0.4351
## Neg Pred Value 0.910485 0.96176 0.8801 0.9606
## Prevalence 0.097059 0.03824 0.1199 0.2605
## Detection Rate 0.008333 0.00000 0.0000 0.2431
## Detection Prevalence 0.008824 0.00000 0.0000 0.5588
## Balanced Accuracy 0.542658 0.50000 0.5000 0.7531
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.63841 0.9095 0.8671
## Specificity 0.99515 0.9942 0.9635
## Pos Pred Value 0.95596 0.9651 0.8509
## Neg Pred Value 0.94342 0.9843 0.9679
## Prevalence 0.14167 0.1490 0.1936
## Detection Rate 0.09044 0.1355 0.1679
## Detection Prevalence 0.09461 0.1404 0.1973
## Balanced Accuracy 0.81678 0.9519 0.9153
ad_tda_kde_5.50.5_n3_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 34 0 0 0 0 0 2
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 347 156 475 992 184 30 96
## HOROZ 3 0 9 0 369 0 5
## SEKER 1 0 0 17 0 553 2
## SIRA 11 0 5 54 25 25 685
##
## Overall Statistics
##
## Accuracy : 0.6453
## 95% CI : (0.6304, 0.66)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5459
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.085859 0.00000 0.0000 0.9332
## Specificity 0.999457 1.00000 1.0000 0.5731
## Pos Pred Value 0.944444 NaN NaN 0.4351
## Neg Pred Value 0.910485 0.96176 0.8801 0.9606
## Prevalence 0.097059 0.03824 0.1199 0.2605
## Detection Rate 0.008333 0.00000 0.0000 0.2431
## Detection Prevalence 0.008824 0.00000 0.0000 0.5588
## Balanced Accuracy 0.542658 0.50000 0.5000 0.7531
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.63841 0.9095 0.8671
## Specificity 0.99515 0.9942 0.9635
## Pos Pred Value 0.95596 0.9651 0.8509
## Neg Pred Value 0.94342 0.9843 0.9679
## Prevalence 0.14167 0.1490 0.1936
## Detection Rate 0.09044 0.1355 0.1679
## Detection Prevalence 0.09461 0.1404 0.1973
## Balanced Accuracy 0.81678 0.9519 0.9153
ad_tda_kde_5.50.5_n3_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6453431 0.5459017 0.6304409 0.6600370 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
ad_tda_kde_5.50.5_n3_db_svm_cf0_ov_acc<-ad_tda_kde_5.50.5_n3_db_svm_cf0$overall[1]
ad_tda_kde_5.50.5_n3_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.08585859 0.9994571 0.9444444 0.9104847 0.9444444
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.00000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.93320790 0.5730858 0.4350877 0.9605556 0.4350877
## Class: HOROZ 0.63840830 0.9951456 0.9559585 0.9434218 0.9559585
## Class: SEKER 0.90953947 0.9942396 0.9650960 0.9843171 0.9650960
## Class: SIRA 0.86708861 0.9635258 0.8509317 0.9679389 0.8509317
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.08585859 0.1574074 0.09705882 0.008333333
## Class: BOMBAY 0.00000000 NA 0.03823529 0.000000000
## Class: CALI 0.00000000 NA 0.11985294 0.000000000
## Class: DERMASON 0.93320790 0.5934789 0.26053922 0.243137255
## Class: HOROZ 0.63840830 0.7655602 0.14166667 0.090441176
## Class: SEKER 0.90953947 0.9364945 0.14901961 0.135539216
## Class: SIRA 0.86708861 0.8589342 0.19362745 0.167892157
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.008823529 0.5426578
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000000000 0.5000000
## Class: DERMASON 0.558823529 0.7531469
## Class: HOROZ 0.094607843 0.8167770
## Class: SEKER 0.140441176 0.9518896
## Class: SIRA 0.197303922 0.9153072
ad_tda_kde_5.50.5_n3_db_svm_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n3_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_svm_n3_3_fold<-(db_svm_fit_re - ad_tda_kde_5.50.5_n3_svm_fit_re)
diff_drybean_tda_kde_5.50.5_svm_n3_3_fold
## Accuracy
## 1 0.001275502
## 2 0.011491100
## 3 0.021650084
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n3_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n3_3_fold_odds.left<-bst_tda_kde_5.50.5_svm.n3_3_fold$probLeft/bst_tda_kde_5.50.5_svm.n3_3_fold$probRight
bst_tda_kde_5.50.5_svm.n3_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n3_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.394
##
## $winRight
## [1] 0.606
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n3_3_fold
## $left
## [1] 0.04358046
##
## $rope
## [1] 0.3806632
##
## $right
## [1] 0.5757564
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold))
#bf_tda_kde_5.50.5_svm.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_svm_n3_3_fold)
## t = 1.9505, df = 2, p-value = 0.1904
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.01383444 0.03677889
## sample estimates:
## mean of x
## 0.01147223
### Test set diff
diff_drybean_tda_kde_5.50.5_svm.n3_test<-(db_svm_cf_ov_acc-ad_tda_kde_5.50.5_n3_db_svm_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_svm.n3_test
## Accuracy
## 0.2816176
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n3_test),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n3_test_odds.left<-bst_tda_kde_5.50.5_svm.n3_test$probLeft/bst_tda_kde_5.50.5_svm.n3_test$probRight
bst_tda_kde_5.50.5_svm.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n3_test),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1601
##
## $winRight
## [1] 0.8399
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n3_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_svm.n3_test))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm.n3_test)) #bf_tda_kde_5.50.5_svm.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n3_test))
##Node4
DryBean_TDA_KDE_5.50.5_n4_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_KDE_5.50.5_n4_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 2024 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1349, 1348, 1351
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.8176688 0.68953858
## 0.1 0.50 0.8211227 0.69653284
## 0.1 0.75 0.8240834 0.70269504
## 0.1 1.00 0.8221067 0.70002358
## 0.1 1.25 0.8201343 0.69697728
## 1.0 0.25 0.7346624 0.50733591
## 1.0 0.50 0.7687526 0.58599720
## 1.0 0.75 0.7791253 0.61320332
## 1.0 1.00 0.7781449 0.61638940
## 1.0 1.25 0.7786395 0.61975579
## 10.0 0.25 0.5197634 0.00000000
## 10.0 0.50 0.5197634 0.00000000
## 10.0 0.75 0.5227264 0.00769601
## 10.0 1.00 0.5296466 0.02558484
## 10.0 1.25 0.5424883 0.05896260
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 0.75.
DryBean_TDA_KDE_5.50.5_n4_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.8296296 0.7116606 Fold1
## 2 0.8387574 0.7271691 Fold2
## 3 0.8038633 0.6692555 Fold3
ad_tda_kde_5.50.5_n4_svm_fit_re<-DryBean_TDA_KDE_5.50.5_n4_SvmFit0 $resample[1]
summary(DryBean_TDA_KDE_5.50.5_n4_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_KDE_5.50.5_n4_SvmFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n4_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n4_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n4_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n4_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 395 156 489 1006 577 170 528
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 17 0 430 1
## SIRA 1 0 0 40 1 8 261
##
## Overall Statistics
##
## Accuracy : 0.4159
## 95% CI : (0.4007, 0.4312)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2282
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9464
## Specificity 1.00000 1.00000 1.0000 0.2327
## Pos Pred Value NaN NaN NaN 0.3029
## Neg Pred Value 0.90294 0.96176 0.8801 0.9249
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2466
## Detection Prevalence 0.00000 0.00000 0.0000 0.8140
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5895
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.7072 0.33038
## Specificity 1.0000 0.9948 0.98480
## Pos Pred Value NaN 0.9598 0.83923
## Neg Pred Value 0.8583 0.9510 0.85964
## Prevalence 0.1417 0.1490 0.19363
## Detection Rate 0.0000 0.1054 0.06397
## Detection Prevalence 0.0000 0.1098 0.07623
## Balanced Accuracy 0.5000 0.8510 0.65759
ad_tda_kde_5.50.5_n4_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 395 156 489 1006 577 170 528
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 17 0 430 1
## SIRA 1 0 0 40 1 8 261
##
## Overall Statistics
##
## Accuracy : 0.4159
## 95% CI : (0.4007, 0.4312)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2282
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9464
## Specificity 1.00000 1.00000 1.0000 0.2327
## Pos Pred Value NaN NaN NaN 0.3029
## Neg Pred Value 0.90294 0.96176 0.8801 0.9249
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2466
## Detection Prevalence 0.00000 0.00000 0.0000 0.8140
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5895
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.7072 0.33038
## Specificity 1.0000 0.9948 0.98480
## Pos Pred Value NaN 0.9598 0.83923
## Neg Pred Value 0.8583 0.9510 0.85964
## Prevalence 0.1417 0.1490 0.19363
## Detection Rate 0.0000 0.1054 0.06397
## Detection Prevalence 0.0000 0.1098 0.07623
## Balanced Accuracy 0.5000 0.8510 0.65759
ad_tda_kde_5.50.5_n4_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.159314e-01 2.282457e-01 4.007498e-01 4.312335e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 6.264481e-103 NaN
ad_tda_kde_5.50.5_n4_db_svm_cf0_ov_acc<-ad_tda_kde_5.50.5_n4_db_svm_cf0$overall[1]
ad_tda_kde_5.50.5_n4_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.9463782 0.2326815 0.3029208 0.9249012 0.3029208
## Class: HOROZ 0.0000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.7072368 0.9948157 0.9598214 0.9509912 0.9598214
## Class: SIRA 0.3303797 0.9848024 0.8392283 0.8596445 0.8392283
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.00000000
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.0000000 NA 0.11985294 0.00000000
## Class: DERMASON 0.9463782 0.4589416 0.26053922 0.24656863
## Class: HOROZ 0.0000000 NA 0.14166667 0.00000000
## Class: SEKER 0.7072368 0.8143939 0.14901961 0.10539216
## Class: SIRA 0.3303797 0.4741144 0.19362745 0.06397059
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.00000000 0.5000000
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.00000000 0.5000000
## Class: DERMASON 0.81397059 0.5895298
## Class: HOROZ 0.00000000 0.5000000
## Class: SEKER 0.10980392 0.8510263
## Class: SIRA 0.07622549 0.6575911
ad_tda_kde_5.50.5_n4_db_svm_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n4_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_svm_n4_3_fold<-(db_svm_fit_re - ad_tda_kde_5.50.5_n4_svm_fit_re)
diff_drybean_tda_kde_5.50.5_svm_n4_3_fold
## Accuracy
## 1 0.09422185
## 2 0.09325393
## 3 0.12812663
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n4_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n4_3_fold_odds.left<-bst_tda_kde_5.50.5_svm.n4_3_fold$probLeft/bst_tda_kde_5.50.5_svm.n4_3_fold$probRight
bst_tda_kde_5.50.5_svm.n4_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n4_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.0084
##
## $winRight
## [1] 0.9916
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n4_3_fold
## $left
## [1] 0.006476539
##
## $rope
## [1] 0.002922708
##
## $right
## [1] 0.9906008
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold))
#bf_tda_kde_5.50.5_svm.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_svm_n4_3_fold)
## t = 9.1748, df = 2, p-value = 0.01167
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.05586523 0.15453638
## sample estimates:
## mean of x
## 0.1052008
### Test set diff
diff_drybean_tda_kde_5.50.5_svm.n4_test<-(db_svm_cf_ov_acc-ad_tda_kde_5.50.5_n4_db_svm_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_svm.n4_test
## Accuracy
## 0.5110294
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n4_test_odds.left<-bst_tda_kde_5.50.5_svm.n4_test$probLeft/bst_tda_kde_5.50.5_svm.n4_test$probRight
bst_tda_kde_5.50.5_svm.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1603667
##
## $winRight
## [1] 0.8396333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_svm.n4_test))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test)) #bf_tda_kde_5.50.5_svm.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test))
##Node5
DryBean_TDA_KDE_5.50.5_n5_SvmFit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'svmRadial',
trControl = fitControl,
tuneGrid = svmGrid, preProc = c('center','scale'),
metric='Accuracy')
DryBean_TDA_KDE_5.50.5_n5_SvmFit0
## Support Vector Machines with Radial Basis Function Kernel
##
## 989 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## Pre-processing: centered (16), scaled (16)
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 659, 660, 659
## Resampling results across tuning parameters:
##
## sigma C Accuracy Kappa
## 0.1 0.25 0.7249762 0.470647847
## 0.1 0.50 0.7492278 0.531246300
## 0.1 0.75 0.7482239 0.532952329
## 0.1 1.00 0.7492340 0.537413770
## 0.1 1.25 0.7542845 0.550066092
## 1.0 0.25 0.6248749 0.187999649
## 1.0 0.50 0.6643087 0.304622052
## 1.0 0.75 0.6835191 0.361151065
## 1.0 1.00 0.6835129 0.380201828
## 1.0 1.25 0.6966381 0.422078321
## 10.0 0.25 0.5621842 0.000000000
## 10.0 0.50 0.5621842 0.000000000
## 10.0 0.75 0.5621842 0.000000000
## 10.0 1.00 0.5621842 0.000000000
## 10.0 1.25 0.5652175 0.009307213
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were sigma = 0.1 and C = 1.25.
DryBean_TDA_KDE_5.50.5_n5_SvmFit0$resample
## Accuracy Kappa Resample
## 1 0.7666667 0.5689714 Fold1
## 2 0.7545455 0.5571864 Fold3
## 3 0.7416413 0.5240405 Fold2
ad_tda_kde_5.50.5_n5_svm_fit_re<-DryBean_TDA_KDE_5.50.5_n5_SvmFit0 $resample[1]
summary(DryBean_TDA_KDE_5.50.5_n5_SvmFit0)
## Length Class Mode
## 1 ksvm S4
#vip(DryBean_TDA_KDE_5.50.5_n5_SvmFit0,25) + ggtitle("DryBean_TDA_KDE_5.50.5_n5_SvmFit TDA-Assited Svm")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n5_SvmFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n5_SvmFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
ad_tda_kde_5.50.5_n5_db_svm_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
ad_tda_kde_5.50.5_n5_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 396 156 489 1020 577 400 675
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 11 0 202 0
## SIRA 0 0 0 32 1 6 115
##
## Overall Statistics
##
## Accuracy : 0.3277
## 95% CI : (0.3133, 0.3423)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.101
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9595
## Specificity 1.00000 1.00000 1.0000 0.1074
## Pos Pred Value NaN NaN NaN 0.2747
## Neg Pred Value 0.90294 0.96176 0.8801 0.8828
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2500
## Detection Prevalence 0.00000 0.00000 0.0000 0.9100
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5335
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.33224 0.14557
## Specificity 1.0000 0.99683 0.98815
## Pos Pred Value NaN 0.94836 0.74675
## Neg Pred Value 0.8583 0.89501 0.82807
## Prevalence 0.1417 0.14902 0.19363
## Detection Rate 0.0000 0.04951 0.02819
## Detection Prevalence 0.0000 0.05221 0.03775
## Balanced Accuracy 0.5000 0.66453 0.56686
ad_tda_kde_5.50.5_n5_db_svm_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 396 156 489 1020 577 400 675
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 11 0 202 0
## SIRA 0 0 0 32 1 6 115
##
## Overall Statistics
##
## Accuracy : 0.3277
## 95% CI : (0.3133, 0.3423)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.101
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9595
## Specificity 1.00000 1.00000 1.0000 0.1074
## Pos Pred Value NaN NaN NaN 0.2747
## Neg Pred Value 0.90294 0.96176 0.8801 0.8828
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2500
## Detection Prevalence 0.00000 0.00000 0.0000 0.9100
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5335
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.33224 0.14557
## Specificity 1.0000 0.99683 0.98815
## Pos Pred Value NaN 0.94836 0.74675
## Neg Pred Value 0.8583 0.89501 0.82807
## Prevalence 0.1417 0.14902 0.19363
## Detection Rate 0.0000 0.04951 0.02819
## Detection Prevalence 0.0000 0.05221 0.03775
## Balanced Accuracy 0.5000 0.66453 0.56686
ad_tda_kde_5.50.5_n5_db_svm_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.276961e-01 1.009676e-01 3.132987e-01 3.423406e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 8.635197e-22 NaN
ad_tda_kde_5.50.5_n5_db_svm_cf0_ov_acc<-ad_tda_kde_5.50.5_n5_db_svm_cf0$overall[1]
ad_tda_kde_5.50.5_n5_db_svm_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.9595484 0.1073914 0.2747105 0.8828338 0.2747105
## Class: HOROZ 0.0000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.3322368 0.9968318 0.9483568 0.8950091 0.9483568
## Class: SIRA 0.1455696 0.9881459 0.7467532 0.8280693 0.7467532
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.00000000
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.0000000 NA 0.11985294 0.00000000
## Class: DERMASON 0.9595484 0.4271357 0.26053922 0.25000000
## Class: HOROZ 0.0000000 NA 0.14166667 0.00000000
## Class: SEKER 0.3322368 0.4920828 0.14901961 0.04950980
## Class: SIRA 0.1455696 0.2436441 0.19362745 0.02818627
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.00000000 0.5000000
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.00000000 0.5000000
## Class: DERMASON 0.91004902 0.5334699
## Class: HOROZ 0.00000000 0.5000000
## Class: SEKER 0.05220588 0.6645343
## Class: SIRA 0.03774510 0.5668578
ad_tda_kde_5.50.5_n5_db_svm_cf0_pre_rec_f1<-ad_tda_kde_5.50.5_n5_db_svm_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_svm_n5_3_fold<-(db_svm_fit_re - ad_tda_kde_5.50.5_n5_svm_fit_re)
diff_drybean_tda_kde_5.50.5_svm_n5_3_fold
## Accuracy
## 1 0.1571848
## 2 0.1774659
## 3 0.1903486
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n5_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n5_3_fold_odds.left<-bst_tda_kde_5.50.5_svm.n5_3_fold$probLeft/bst_tda_kde_5.50.5_svm.n5_3_fold$probRight
bst_tda_kde_5.50.5_svm.n5_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n5_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.008966667
##
## $winRight
## [1] 0.9910333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n5_3_fold
## $left
## [1] 0.001805102
##
## $rope
## [1] 0.0004609715
##
## $right
## [1] 0.9977339
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold))
#bf_tda_kde_5.50.5_svm.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_svm_n5_3_fold)
## t = 18.13, df = 2, p-value = 0.003029
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.1334678 0.2165317
## sample estimates:
## mean of x
## 0.1749998
### Test set diff
diff_drybean_tda_kde_5.50.5_svm.n5_test<-(db_svm_cf_ov_acc-ad_tda_kde_5.50.5_n5_db_svm_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_svm.n5_test
## Accuracy
## 0.5992647
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n5_test),-0.01,0.01)
bst_tda_kde_5.50.5_svm.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_svm.n5_test_odds.left<-bst_tda_kde_5.50.5_svm.n5_test$probLeft/bst_tda_kde_5.50.5_svm.n5_test$probRight
bst_tda_kde_5.50.5_svm.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_svm.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n5_test),-0.01,0.01)
bsr_tda_kde_5.50.5_svm.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1564333
##
## $winRight
## [1] 0.8435667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_svm.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n5_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_svm.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_svm.n5_test))
#BayesFactor
#bf_tda_kde_5.50.5_svm.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_svm.n5_test)) #bf_tda_kde_5.50.5_svm.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_svm.n4_test))
#Non-TDA-Assisted
nn1Grid<-expand.grid(size = c(2,3,5,7), decay = c(0.3,0.5,0.7))
#Neural Network
dryBeanNn1Fit <- train(as.factor(Class) ~ ., data = Dry_Bean_DatasetTrain,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 55
## initial value 14381.499891
## iter 10 value 11654.878422
## final value 11654.867658
## converged
## # weights: 79
## initial value 13954.918828
## iter 10 value 11658.024779
## iter 20 value 11593.114330
## iter 30 value 10532.207889
## iter 40 value 10283.884833
## iter 50 value 7754.501277
## iter 60 value 7498.477578
## iter 70 value 7097.931381
## iter 80 value 6714.281624
## iter 90 value 5980.868315
## iter 100 value 5767.764361
## final value 5767.764361
## stopped after 100 iterations
## # weights: 127
## initial value 15206.528540
## iter 10 value 11870.803581
## iter 20 value 11655.514193
## iter 30 value 11655.450163
## iter 40 value 11631.196435
## iter 50 value 10760.537668
## iter 60 value 10365.593101
## iter 70 value 8937.080582
## iter 80 value 8204.975592
## iter 90 value 7358.785659
## iter 100 value 6701.362613
## final value 6701.362613
## stopped after 100 iterations
## # weights: 175
## initial value 13286.730083
## iter 10 value 11680.146740
## iter 20 value 11655.227526
## final value 11655.218856
## converged
## # weights: 55
## initial value 13248.623370
## iter 10 value 11677.676269
## iter 20 value 11177.067941
## iter 30 value 10142.756346
## iter 40 value 9533.707395
## iter 50 value 9511.170843
## iter 60 value 9368.549716
## iter 70 value 9112.272827
## iter 80 value 8052.142597
## iter 90 value 7195.761897
## iter 100 value 6389.738490
## final value 6389.738490
## stopped after 100 iterations
## # weights: 79
## initial value 12618.500019
## iter 10 value 11656.290354
## iter 20 value 11655.896866
## iter 30 value 11028.819668
## iter 40 value 10692.778766
## iter 50 value 10580.104115
## iter 60 value 9588.831837
## iter 70 value 9083.853303
## iter 80 value 8877.307848
## iter 90 value 8649.548027
## iter 100 value 8589.135377
## final value 8589.135377
## stopped after 100 iterations
## # weights: 127
## initial value 12557.694008
## iter 10 value 11676.527515
## iter 20 value 11589.902354
## iter 30 value 11487.197145
## iter 40 value 10764.019514
## iter 50 value 10311.908294
## iter 60 value 8614.176076
## iter 70 value 7292.663924
## iter 80 value 5334.693722
## iter 90 value 4617.473909
## iter 100 value 4229.778307
## final value 4229.778307
## stopped after 100 iterations
## # weights: 175
## initial value 13951.466953
## iter 10 value 11664.503686
## iter 20 value 11654.007382
## iter 30 value 9781.819674
## iter 40 value 9005.104611
## iter 50 value 8750.913134
## iter 60 value 7583.329944
## iter 70 value 6988.274472
## iter 80 value 6839.051589
## iter 90 value 6716.240813
## iter 100 value 5482.775766
## final value 5482.775766
## stopped after 100 iterations
## # weights: 55
## initial value 12417.580191
## iter 10 value 11669.136630
## iter 20 value 11655.772180
## iter 30 value 11508.587504
## iter 40 value 11366.854826
## iter 50 value 11151.706526
## iter 60 value 11092.255422
## iter 70 value 10950.511833
## iter 80 value 10733.143807
## iter 90 value 10712.414021
## iter 100 value 10703.818129
## final value 10703.818129
## stopped after 100 iterations
## # weights: 79
## initial value 13779.154394
## iter 10 value 11657.498248
## iter 20 value 11074.637359
## iter 30 value 9414.582853
## iter 40 value 8844.186004
## iter 50 value 8592.434280
## iter 60 value 8404.480606
## iter 70 value 8152.389331
## iter 80 value 8102.460795
## iter 90 value 7583.569279
## iter 100 value 6791.588951
## final value 6791.588951
## stopped after 100 iterations
## # weights: 127
## initial value 13415.416000
## iter 10 value 11657.424549
## iter 20 value 11655.379221
## iter 30 value 11655.163758
## iter 40 value 11654.925699
## final value 11654.922575
## converged
## # weights: 175
## initial value 13590.414173
## iter 10 value 11655.058426
## iter 20 value 11654.901368
## iter 30 value 9774.304635
## iter 40 value 9342.921751
## iter 50 value 9224.001780
## iter 60 value 8992.371996
## iter 70 value 7962.682061
## iter 80 value 7273.385903
## iter 90 value 7089.651309
## iter 100 value 6734.843636
## final value 6734.843636
## stopped after 100 iterations
## # weights: 55
## initial value 13115.576105
## iter 10 value 11657.167604
## iter 20 value 10968.561794
## iter 30 value 10292.360687
## iter 40 value 9261.587631
## iter 50 value 7840.337705
## iter 60 value 7015.055920
## iter 70 value 6678.699638
## iter 80 value 6093.633016
## iter 90 value 5495.839193
## iter 100 value 5392.100169
## final value 5392.100169
## stopped after 100 iterations
## # weights: 79
## initial value 13275.747067
## iter 10 value 11657.330695
## final value 11656.990589
## converged
## # weights: 127
## initial value 15017.520089
## iter 10 value 11656.935101
## final value 11656.934882
## converged
## # weights: 175
## initial value 13502.670617
## iter 10 value 11657.040715
## iter 20 value 11655.631599
## iter 20 value 11655.631487
## iter 30 value 11433.411930
## iter 40 value 11395.305527
## iter 50 value 11388.528113
## iter 60 value 10089.307659
## iter 70 value 9785.962916
## iter 80 value 8899.492790
## iter 90 value 8742.089500
## iter 100 value 8703.726457
## final value 8703.726457
## stopped after 100 iterations
## # weights: 55
## initial value 12215.159457
## iter 10 value 11658.333553
## iter 20 value 11646.129051
## iter 30 value 9857.041891
## iter 40 value 9470.943949
## iter 50 value 8615.932731
## iter 60 value 8064.960716
## iter 70 value 7837.799899
## iter 80 value 7472.960803
## iter 90 value 7055.443348
## iter 100 value 6828.698642
## final value 6828.698642
## stopped after 100 iterations
## # weights: 79
## initial value 12996.067957
## iter 10 value 11665.950407
## iter 20 value 11657.253830
## iter 30 value 11657.140565
## final value 11657.139252
## converged
## # weights: 127
## initial value 12302.186601
## iter 10 value 11660.789717
## iter 20 value 11657.624723
## iter 30 value 11607.657716
## iter 40 value 11003.840530
## iter 50 value 10837.980447
## iter 60 value 10590.204814
## iter 70 value 10030.588648
## iter 80 value 8837.725183
## iter 90 value 8342.075079
## iter 100 value 7839.172144
## final value 7839.172144
## stopped after 100 iterations
## # weights: 175
## initial value 12935.519038
## iter 10 value 11376.664909
## iter 20 value 10359.335841
## iter 30 value 9089.111676
## iter 40 value 8737.015286
## iter 50 value 8677.405050
## iter 60 value 7568.373198
## iter 70 value 7412.116397
## iter 80 value 6985.874266
## iter 90 value 6887.547524
## iter 100 value 6829.672350
## final value 6829.672350
## stopped after 100 iterations
## # weights: 55
## initial value 12609.733759
## iter 10 value 11658.831358
## iter 20 value 10254.131588
## iter 30 value 10126.286785
## iter 40 value 10078.256609
## iter 50 value 10019.529517
## iter 60 value 9986.144781
## final value 9975.862831
## converged
## # weights: 79
## initial value 12832.486592
## iter 10 value 11682.830808
## final value 11657.287941
## converged
## # weights: 127
## initial value 18274.661880
## iter 10 value 11738.793043
## iter 20 value 11658.355078
## iter 30 value 10139.244822
## iter 40 value 9035.254868
## iter 50 value 8593.689987
## iter 60 value 8246.535074
## iter 70 value 8043.386442
## iter 80 value 7897.369561
## iter 90 value 7489.492726
## iter 100 value 7209.193773
## final value 7209.193773
## stopped after 100 iterations
## # weights: 175
## initial value 13903.517617
## iter 10 value 11671.760580
## iter 20 value 11657.509604
## iter 30 value 11657.290174
## final value 11657.288140
## converged
## # weights: 55
## initial value 15344.925451
## iter 10 value 11693.835490
## iter 20 value 11659.513960
## iter 30 value 11032.757223
## iter 40 value 10796.974804
## iter 50 value 10753.907202
## iter 60 value 10716.116637
## iter 70 value 10466.769551
## iter 80 value 8851.623071
## iter 90 value 6597.787585
## iter 100 value 6360.056445
## final value 6360.056445
## stopped after 100 iterations
## # weights: 79
## initial value 12203.359891
## iter 10 value 11688.003336
## iter 20 value 11659.542838
## iter 30 value 11147.990696
## iter 40 value 10700.342014
## iter 50 value 10028.156638
## iter 60 value 9246.224076
## iter 70 value 8375.221409
## iter 80 value 7427.580750
## iter 90 value 6781.929680
## iter 100 value 5365.301377
## final value 5365.301377
## stopped after 100 iterations
## # weights: 127
## initial value 15178.295692
## iter 10 value 11787.994268
## iter 20 value 11659.713952
## iter 30 value 11658.969958
## iter 30 value 11658.969929
## iter 30 value 11658.969817
## final value 11658.969817
## converged
## # weights: 175
## initial value 15573.083516
## iter 10 value 11783.876871
## iter 20 value 11272.961614
## iter 30 value 9002.217396
## iter 40 value 8901.417888
## iter 50 value 8392.003332
## iter 60 value 7521.789049
## iter 70 value 6360.603822
## iter 80 value 5786.129495
## iter 90 value 5498.048810
## iter 100 value 4973.120595
## final value 4973.120595
## stopped after 100 iterations
## # weights: 55
## initial value 12013.574905
## iter 10 value 11668.772616
## iter 20 value 11655.080512
## iter 30 value 10851.696619
## iter 40 value 10323.750808
## iter 50 value 8734.819267
## iter 60 value 8532.713078
## iter 70 value 8471.685866
## iter 80 value 8284.631796
## iter 90 value 6597.283020
## iter 100 value 5097.134754
## final value 5097.134754
## stopped after 100 iterations
## # weights: 79
## initial value 13651.791329
## iter 10 value 11665.874518
## iter 20 value 11659.263896
## iter 30 value 11659.108410
## iter 40 value 10214.208258
## iter 50 value 9929.416188
## iter 60 value 9749.931163
## iter 70 value 9258.185412
## iter 80 value 9055.524153
## iter 90 value 8750.318558
## iter 100 value 8351.209424
## final value 8351.209424
## stopped after 100 iterations
## # weights: 127
## initial value 12909.373007
## iter 10 value 11664.194883
## iter 20 value 11659.135143
## iter 30 value 11659.081599
## iter 40 value 11563.367156
## iter 50 value 10967.423138
## iter 60 value 9855.017847
## iter 70 value 9582.123202
## iter 80 value 8453.411696
## iter 90 value 8368.701710
## iter 100 value 8317.968660
## final value 8317.968660
## stopped after 100 iterations
## # weights: 175
## initial value 12961.940992
## iter 10 value 11663.929692
## iter 10 value 11663.929618
## iter 20 value 11661.074598
## iter 30 value 11327.264936
## iter 40 value 10535.846453
## iter 50 value 10017.905173
## iter 60 value 8930.639048
## iter 70 value 8272.818289
## iter 80 value 8218.580658
## iter 90 value 7666.977106
## iter 100 value 7063.258615
## final value 7063.258615
## stopped after 100 iterations
## # weights: 55
## initial value 13851.354348
## iter 10 value 11677.821182
## iter 20 value 11661.288894
## iter 30 value 11612.804883
## iter 40 value 9673.142277
## iter 50 value 9168.805815
## iter 60 value 8168.576176
## iter 70 value 7076.154833
## iter 80 value 5889.272526
## iter 90 value 5793.180004
## iter 100 value 5411.723020
## final value 5411.723020
## stopped after 100 iterations
## # weights: 79
## initial value 14079.154110
## iter 10 value 11746.915859
## iter 20 value 11656.406454
## iter 30 value 11092.043448
## iter 40 value 10800.008619
## iter 50 value 9266.778803
## iter 60 value 8790.602969
## iter 70 value 8571.328929
## iter 80 value 8358.772422
## iter 90 value 8166.052165
## iter 100 value 8053.344176
## final value 8053.344176
## stopped after 100 iterations
## # weights: 127
## initial value 13262.985242
## iter 10 value 11659.374176
## final value 11659.192371
## converged
## # weights: 175
## initial value 14400.233092
## iter 10 value 11766.267424
## iter 20 value 11604.522865
## iter 30 value 11270.014206
## iter 40 value 10232.276697
## iter 50 value 9388.611955
## iter 60 value 8773.033610
## iter 70 value 7627.954185
## iter 80 value 7404.640214
## iter 90 value 5856.326119
## iter 100 value 5076.978649
## final value 5076.978649
## stopped after 100 iterations
## # weights: 55
## initial value 19525.104548
## iter 10 value 17488.829220
## iter 20 value 16837.277549
## iter 30 value 15399.635445
## iter 40 value 13724.063716
## iter 50 value 13475.957473
## iter 60 value 12891.857172
## iter 70 value 11581.421930
## iter 80 value 10834.335539
## iter 90 value 10384.890659
## iter 100 value 10367.355463
## final value 10367.355463
## stopped after 100 iterations
dryBeanNn1Fit
## Neural Network
##
## 9531 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6353, 6354, 6355
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.4861303 0.3375035
## 2 0.5 0.6020500 0.5083878
## 2 0.7 0.4295806 0.2673673
## 3 0.3 0.5459169 0.4127435
## 3 0.5 0.3694305 0.1846085
## 3 0.7 0.3857824 0.2053809
## 5 0.3 0.3630985 0.1554784
## 5 0.5 0.5563549 0.4406289
## 5 0.7 0.3484419 0.1399181
## 7 0.3 0.4478486 0.2802325
## 7 0.5 0.5782216 0.4778784
## 7 0.7 0.5055201 0.3676189
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 2 and decay = 0.5.
dryBeanNn1Fit$resample
## Accuracy Kappa Resample
## 1 0.7056045 0.6355485 Fold3
## 2 0.5332074 0.4213065 Fold2
## 3 0.5673379 0.4683083 Fold1
db_nn1_fit_re<-dryBeanNn1Fit$resample[1]
summary(dryBeanNn1Fit)
## a 16-2-7 network with 55 weights
## options were - softmax modelling decay=0.5
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## -0.28 0.00 -0.01 0.36 0.42 -4.40 7.02 0.00 -0.66 -1.53
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## -0.68 -0.83 -0.14 -0.04 -0.01 -1.06 -0.66
## b->o1 h1->o1 h2->o1
## 4.96 0.09 -11.66
## b->o2 h1->o2 h2->o2
## 4.20 0.21 -17.78
## b->o3 h1->o3 h2->o3
## 3.70 0.06 -3.23
## b->o4 h1->o4 h2->o4
## -0.91 -0.48 7.36
## b->o5 h1->o5 h2->o5
## -16.43 0.16 24.05
## b->o6 h1->o6 h2->o6
## 3.43 0.50 -3.03
## b->o7 h1->o7 h2->o7
## 1.03 -0.54 4.27
#vip(dryBeanNn1Fit,25) + ggtitle("non-TDA-Assited NN")
# Predict outcome using model from training data based on testing data
predictions <- predict(dryBeanNn1Fit, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_nn1_cf<-confusionMatrix(data=predictions, as.factor(Dry_Bean_DatasetTest$Class))
db_nn1_cf
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 308 156 174 0 0 19 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 1 0 0 0 0
## DERMASON 4 0 32 919 41 3 484
## HOROZ 0 0 3 49 527 0 20
## SEKER 77 0 241 5 3 508 61
## SIRA 7 0 38 90 7 78 225
##
## Overall Statistics
##
## Accuracy : 0.6098
## 95% CI : (0.5946, 0.6248)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5213
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.77778 0.00000 0.0020450 0.8645
## Specificity 0.90527 1.00000 1.0000000 0.8131
## Pos Pred Value 0.46880 NaN 1.0000000 0.6197
## Neg Pred Value 0.97429 0.96176 0.8803628 0.9446
## Prevalence 0.09706 0.03824 0.1198529 0.2605
## Detection Rate 0.07549 0.00000 0.0002451 0.2252
## Detection Prevalence 0.16103 0.00000 0.0002451 0.3635
## Balanced Accuracy 0.84152 0.50000 0.5010225 0.8388
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9118 0.8355 0.28481
## Specificity 0.9794 0.8885 0.93313
## Pos Pred Value 0.8798 0.5676 0.50562
## Neg Pred Value 0.9853 0.9686 0.84457
## Prevalence 0.1417 0.1490 0.19363
## Detection Rate 0.1292 0.1245 0.05515
## Detection Prevalence 0.1468 0.2194 0.10907
## Balanced Accuracy 0.9456 0.8620 0.60897
db_nn1_cf$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6098039 0.5212516 0.5946394 0.6248110 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_nn1_cf_ov_acc<-db_nn1_cf$overall[1]
db_nn1_cf$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.77777778 0.9052660 0.4687976 0.9742916 0.4687976
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.00204499 1.0000000 1.0000000 0.8803628 1.0000000
## Class: DERMASON 0.86453434 0.8130593 0.6196898 0.9445514 0.6196898
## Class: HOROZ 0.91176471 0.9794403 0.8797997 0.9853490 0.8797997
## Class: SEKER 0.83552632 0.8885369 0.5675978 0.9686028 0.5675978
## Class: SIRA 0.28481013 0.9331307 0.5056180 0.8445667 0.5056180
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.77777778 0.584995252 0.09705882 0.075490196
## Class: BOMBAY 0.00000000 NA 0.03823529 0.000000000
## Class: CALI 0.00204499 0.004081633 0.11985294 0.000245098
## Class: DERMASON 0.86453434 0.721916732 0.26053922 0.225245098
## Class: HOROZ 0.91176471 0.895497026 0.14166667 0.129166667
## Class: SEKER 0.83552632 0.675981371 0.14901961 0.124509804
## Class: SIRA 0.28481013 0.364372470 0.19362745 0.055147059
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.161029412 0.8415219
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000245098 0.5010225
## Class: DERMASON 0.363480392 0.8387968
## Class: HOROZ 0.146813725 0.9456025
## Class: SEKER 0.219362745 0.8620316
## Class: SIRA 0.109068627 0.6089704
db_nn1_cf_pre_rec_f1<-db_nn1_cf$byClass[5:7]
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node1
#Neural Network 1
DryBean_TDA_PC_5.50.5_n1_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 52
## initial value 9387.117698
## iter 10 value 5849.075506
## iter 20 value 5844.997033
## iter 30 value 5826.635205
## iter 40 value 5278.069745
## iter 50 value 4449.761436
## iter 60 value 4381.947117
## iter 70 value 4321.963534
## iter 80 value 4255.482653
## iter 90 value 4215.830160
## iter 100 value 4148.376629
## final value 4148.376629
## stopped after 100 iterations
## # weights: 75
## initial value 9102.360718
## iter 10 value 5964.402837
## iter 20 value 5847.551152
## iter 30 value 5844.366499
## iter 40 value 5842.794383
## iter 50 value 5842.411745
## iter 60 value 4481.617706
## iter 70 value 3940.091992
## iter 80 value 3798.573930
## iter 90 value 3558.920024
## iter 100 value 2929.779277
## final value 2929.779277
## stopped after 100 iterations
## # weights: 121
## initial value 10717.577008
## iter 10 value 5882.507082
## iter 20 value 5847.847160
## iter 30 value 5807.585868
## iter 40 value 5140.668782
## iter 50 value 4216.980457
## iter 60 value 2788.065252
## iter 70 value 2374.969830
## iter 80 value 2237.413688
## iter 90 value 2024.863693
## iter 100 value 1715.440075
## final value 1715.440075
## stopped after 100 iterations
## # weights: 167
## initial value 10256.434164
## iter 10 value 5871.324644
## iter 20 value 5845.218734
## iter 30 value 5841.451211
## iter 40 value 5841.343354
## iter 50 value 5841.178938
## iter 60 value 5838.044347
## iter 70 value 5827.311221
## iter 80 value 5813.127256
## iter 90 value 5342.447321
## iter 100 value 4009.725776
## final value 4009.725776
## stopped after 100 iterations
## # weights: 52
## initial value 8243.357291
## iter 10 value 5974.967665
## iter 20 value 5853.715008
## iter 30 value 5846.398911
## iter 40 value 5845.905530
## final value 5845.900460
## converged
## # weights: 75
## initial value 9748.981100
## iter 10 value 5999.739218
## iter 20 value 5853.238185
## iter 30 value 5849.315281
## iter 40 value 5848.764223
## iter 50 value 5845.904781
## final value 5845.900240
## converged
## # weights: 121
## initial value 9996.167283
## iter 10 value 6008.771215
## iter 20 value 5857.854433
## iter 30 value 5756.375324
## iter 40 value 4096.332066
## iter 50 value 3813.150466
## iter 60 value 3579.681747
## iter 70 value 2645.122775
## iter 80 value 2268.195362
## iter 90 value 2155.113932
## iter 100 value 2031.518687
## final value 2031.518687
## stopped after 100 iterations
## # weights: 167
## initial value 17197.406493
## iter 10 value 5869.593965
## iter 20 value 5845.359669
## iter 30 value 5845.054597
## iter 40 value 5514.254930
## iter 50 value 4414.102307
## iter 60 value 4166.676859
## iter 70 value 4057.263781
## iter 80 value 3808.358940
## iter 90 value 3675.037444
## iter 100 value 3613.762015
## final value 3613.762015
## stopped after 100 iterations
## # weights: 52
## initial value 10122.966841
## iter 10 value 5862.316443
## iter 20 value 5848.416875
## final value 5848.383646
## converged
## # weights: 75
## initial value 7806.850901
## iter 10 value 5869.110961
## iter 20 value 5211.046785
## iter 30 value 4046.699316
## iter 40 value 3853.732001
## iter 50 value 3818.179972
## iter 60 value 3563.494616
## iter 70 value 3499.968676
## iter 80 value 3008.125410
## iter 90 value 2743.318048
## iter 100 value 1995.342795
## final value 1995.342795
## stopped after 100 iterations
## # weights: 121
## initial value 14860.450452
## iter 10 value 5916.701228
## iter 20 value 5634.568305
## iter 30 value 4174.133782
## iter 40 value 3529.408919
## iter 50 value 2969.476789
## iter 60 value 2742.889865
## iter 70 value 2354.535187
## iter 80 value 2124.137692
## iter 90 value 2022.571824
## iter 100 value 1870.040140
## final value 1870.040140
## stopped after 100 iterations
## # weights: 167
## initial value 12076.093540
## iter 10 value 6003.278141
## iter 20 value 5533.819720
## iter 30 value 4930.291971
## iter 40 value 4373.451267
## iter 50 value 3627.189425
## iter 60 value 2789.788890
## iter 70 value 2396.423561
## iter 80 value 2172.030415
## iter 90 value 2040.568048
## iter 100 value 1870.144773
## final value 1870.144773
## stopped after 100 iterations
## # weights: 52
## initial value 10463.022489
## iter 10 value 5863.055700
## iter 20 value 5856.588277
## iter 30 value 5851.329770
## iter 40 value 5671.254799
## iter 50 value 5648.558532
## iter 60 value 5591.015616
## iter 70 value 4694.260519
## iter 80 value 4376.218663
## iter 90 value 4108.075762
## iter 100 value 3953.430705
## final value 3953.430705
## stopped after 100 iterations
## # weights: 75
## initial value 9020.658783
## iter 10 value 5880.067360
## iter 20 value 5866.995881
## iter 30 value 5463.716851
## iter 40 value 4134.003599
## iter 50 value 3971.168538
## iter 60 value 3107.582278
## iter 70 value 2600.072920
## iter 80 value 2130.470590
## iter 90 value 1973.305062
## iter 100 value 1867.966523
## final value 1867.966523
## stopped after 100 iterations
## # weights: 121
## initial value 11099.026415
## iter 10 value 5869.951432
## iter 20 value 5851.582799
## iter 30 value 5850.445823
## iter 40 value 5850.364981
## iter 50 value 5850.342453
## iter 60 value 5849.904325
## iter 70 value 5417.240781
## iter 80 value 4861.423086
## iter 90 value 4177.206736
## iter 100 value 3878.577201
## final value 3878.577201
## stopped after 100 iterations
## # weights: 167
## initial value 9409.994182
## iter 10 value 6039.032130
## iter 20 value 5952.826267
## iter 30 value 5771.011198
## iter 40 value 4711.147072
## iter 50 value 4438.077170
## iter 60 value 4420.550102
## iter 70 value 3412.210965
## iter 80 value 3109.320927
## iter 90 value 2968.229430
## iter 100 value 2579.636162
## final value 2579.636162
## stopped after 100 iterations
## # weights: 52
## initial value 9398.585605
## iter 10 value 5892.126452
## iter 20 value 5871.511930
## iter 30 value 5856.158991
## iter 40 value 4591.062928
## iter 50 value 3970.241625
## iter 60 value 3863.799188
## iter 70 value 3831.470180
## iter 80 value 3784.837543
## iter 90 value 3632.817077
## iter 100 value 3430.006087
## final value 3430.006087
## stopped after 100 iterations
## # weights: 75
## initial value 10122.293192
## iter 10 value 5980.165358
## iter 20 value 5860.858083
## iter 30 value 5852.701419
## iter 40 value 5852.685667
## final value 5852.679276
## converged
## # weights: 121
## initial value 7652.413716
## iter 10 value 5927.239489
## iter 20 value 5855.667597
## iter 30 value 5852.316530
## iter 40 value 5851.774045
## final value 5851.759643
## converged
## # weights: 167
## initial value 7238.502817
## iter 10 value 5859.391850
## iter 20 value 5852.511083
## iter 30 value 5557.860980
## iter 40 value 5314.605536
## iter 50 value 5272.880621
## iter 60 value 3234.555813
## iter 70 value 2845.351643
## iter 80 value 2480.063576
## iter 90 value 2371.849228
## iter 100 value 2259.826501
## final value 2259.826501
## stopped after 100 iterations
## # weights: 52
## initial value 7967.442663
## iter 10 value 5880.838883
## iter 20 value 5860.212527
## iter 30 value 5856.615845
## iter 40 value 5856.593842
## iter 40 value 5856.593785
## iter 40 value 5856.593777
## final value 5856.593777
## converged
## # weights: 75
## initial value 7129.999437
## iter 10 value 5861.092505
## iter 20 value 5770.652566
## iter 30 value 5630.482791
## iter 40 value 4085.148452
## iter 50 value 3764.442998
## iter 60 value 3720.422597
## iter 70 value 3532.402406
## iter 80 value 3516.498569
## iter 90 value 3457.720035
## iter 100 value 2773.940172
## final value 2773.940172
## stopped after 100 iterations
## # weights: 121
## initial value 10508.343255
## iter 10 value 5879.895374
## iter 20 value 5860.493780
## iter 30 value 5854.327573
## iter 40 value 5504.010261
## iter 50 value 4496.402024
## iter 60 value 4364.764264
## iter 70 value 4089.055035
## iter 80 value 3599.115623
## iter 90 value 2991.640084
## iter 100 value 2175.543339
## final value 2175.543339
## stopped after 100 iterations
## # weights: 167
## initial value 9326.299622
## iter 10 value 5876.222807
## iter 20 value 5855.680900
## iter 30 value 5853.230720
## iter 40 value 5840.954305
## iter 50 value 5744.205810
## iter 60 value 4694.572628
## iter 70 value 4440.183537
## iter 80 value 3896.862950
## iter 90 value 3227.326330
## iter 100 value 2506.319883
## final value 2506.319883
## stopped after 100 iterations
## # weights: 52
## initial value 10042.275967
## iter 10 value 5938.215179
## iter 20 value 5888.299898
## iter 30 value 5407.665836
## iter 40 value 4145.067162
## iter 50 value 3903.933593
## iter 60 value 3816.841032
## iter 70 value 3481.808028
## iter 80 value 3324.788162
## iter 90 value 3155.302254
## iter 100 value 2969.211329
## final value 2969.211329
## stopped after 100 iterations
## # weights: 75
## initial value 11375.337869
## iter 10 value 5892.898931
## iter 20 value 5842.325330
## iter 30 value 5837.254598
## iter 40 value 5837.058638
## iter 50 value 5150.634522
## iter 60 value 4619.492362
## iter 70 value 4458.514098
## iter 80 value 3922.046604
## iter 90 value 3855.215557
## iter 100 value 3762.086349
## final value 3762.086349
## stopped after 100 iterations
## # weights: 121
## initial value 9791.424989
## iter 10 value 5888.023934
## iter 20 value 5838.057117
## iter 30 value 5836.619337
## final value 5835.477850
## converged
## # weights: 167
## initial value 9422.685067
## iter 10 value 5837.949845
## iter 20 value 5837.649895
## iter 30 value 5837.229408
## iter 40 value 5836.231597
## iter 50 value 4601.883378
## iter 60 value 3890.646967
## iter 70 value 3832.115843
## iter 80 value 3670.289950
## iter 90 value 2819.100182
## iter 100 value 2631.744933
## final value 2631.744933
## stopped after 100 iterations
## # weights: 52
## initial value 11811.002851
## iter 10 value 6071.063202
## iter 20 value 5859.923322
## iter 30 value 5841.999285
## iter 40 value 5839.996110
## iter 50 value 5839.979613
## final value 5839.979417
## converged
## # weights: 75
## initial value 11897.974843
## iter 10 value 5898.214792
## iter 20 value 5884.900256
## iter 30 value 5852.597077
## iter 40 value 5842.054136
## iter 50 value 5838.620148
## iter 60 value 5838.478931
## iter 70 value 5838.408826
## final value 5838.408410
## converged
## # weights: 121
## initial value 12084.191258
## iter 10 value 5956.641824
## iter 20 value 5844.898023
## iter 30 value 5839.143234
## iter 40 value 5724.275284
## iter 50 value 4008.188603
## iter 60 value 3763.112719
## iter 70 value 3689.659589
## iter 80 value 3525.794699
## iter 90 value 3170.287365
## iter 100 value 2858.462569
## final value 2858.462569
## stopped after 100 iterations
## # weights: 167
## initial value 12123.659143
## iter 10 value 5857.038343
## iter 20 value 5839.768283
## iter 30 value 5837.700417
## iter 40 value 4989.288389
## iter 50 value 4648.056741
## iter 60 value 4641.883450
## iter 70 value 4392.959955
## iter 80 value 3821.743112
## iter 90 value 3467.007621
## iter 100 value 2892.949069
## final value 2892.949069
## stopped after 100 iterations
## # weights: 52
## initial value 10306.723599
## iter 10 value 5954.493774
## iter 20 value 5871.330944
## iter 30 value 5815.655803
## iter 40 value 4719.359348
## iter 50 value 4421.528622
## iter 60 value 4395.730756
## iter 70 value 4353.636160
## iter 80 value 4292.460650
## iter 90 value 3858.151659
## iter 100 value 2405.518666
## final value 2405.518666
## stopped after 100 iterations
## # weights: 75
## initial value 9070.684974
## iter 10 value 5882.578305
## iter 20 value 5842.973502
## iter 30 value 5842.515243
## iter 40 value 5842.155569
## iter 50 value 5581.728181
## iter 60 value 3868.401887
## iter 70 value 3644.482697
## iter 80 value 3564.396390
## iter 90 value 3514.190388
## iter 100 value 3497.026498
## final value 3497.026498
## stopped after 100 iterations
## # weights: 121
## initial value 11806.784094
## iter 10 value 5876.274741
## iter 20 value 5840.578567
## iter 30 value 5839.913055
## iter 40 value 5839.640024
## iter 50 value 5565.310952
## iter 60 value 4964.549459
## iter 70 value 4827.487563
## iter 80 value 4645.984119
## iter 90 value 4356.623684
## iter 100 value 3375.563619
## final value 3375.563619
## stopped after 100 iterations
## # weights: 167
## initial value 8788.428376
## iter 10 value 5879.979379
## iter 20 value 5842.449634
## iter 30 value 5840.262955
## iter 40 value 5839.912462
## iter 50 value 5764.248860
## iter 60 value 5327.127750
## iter 70 value 4325.710224
## iter 80 value 3993.068630
## iter 90 value 3783.579745
## iter 100 value 3657.245612
## final value 3657.245612
## stopped after 100 iterations
## # weights: 121
## initial value 15044.822863
## iter 10 value 8789.111098
## iter 20 value 8778.141506
## iter 30 value 8766.295711
## iter 40 value 8765.799133
## iter 50 value 8745.783781
## iter 60 value 8720.463895
## iter 70 value 8697.304162
## iter 80 value 7598.477232
## iter 90 value 6831.987381
## iter 100 value 6599.726231
## final value 6599.726231
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n1_NN1Fit0
## Neural Network
##
## 7839 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5226, 5228, 5224
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.7133311 0.5360189
## 2 0.5 0.5574222 0.2113577
## 2 0.7 0.5962668 0.2744519
## 3 0.3 0.8052540 0.6987787
## 3 0.5 0.4520986 0.0000000
## 3 0.7 0.8048615 0.6877067
## 5 0.3 0.6710616 0.4395232
## 5 0.5 0.7258828 0.5260170
## 5 0.7 0.8644165 0.7915330
## 7 0.3 0.7893953 0.6746336
## 7 0.5 0.7853187 0.6669141
## 7 0.7 0.8378873 0.7536258
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 5 and decay = 0.7.
DryBean_TDA_PC_5.50.5_n1_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.8897819 0.8296020 Fold1
## 2 0.8114723 0.7134838 Fold3
## 3 0.8919954 0.8315131 Fold2
db_tda_pc_5.50.5_n1_nn1_fit_re<-DryBean_TDA_PC_5.50.5_n1_NN1Fit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_NN1Fit0)
## a 16-5-6 network with 121 weights
## options were - softmax modelling decay=0.7
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.13 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.08 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## -0.01 -1.04 -7.46 -2.19 1.85 -0.06 -0.05 1.19 -1.27 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## -0.01 0.00 0.01 0.00 0.00 0.03 -0.02
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1
## -1.19 -1.19 0.10 -1.19 -1.19 3.55
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2
## -0.72 -0.72 -0.02 -0.72 -0.72 0.26
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3
## 1.40 1.40 0.01 1.40 1.40 -3.70
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4
## -0.06 -0.06 0.04 -0.06 -0.06 -3.30
## b->o5 h1->o5 h2->o5 h3->o5 h4->o5 h5->o5
## 0.69 0.69 -0.08 0.68 0.69 -0.07
## b->o6 h1->o6 h2->o6 h3->o6 h4->o6 h5->o6
## -0.12 -0.12 -0.05 -0.12 -0.12 3.25
#vip(DryBean_TDA_PC_5.50.5_n1_NN1Fit0,25) + ggtitle("dryBean_TDA_PCA_5.50.5_n1_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_PC_5.50.5_n1_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n1_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n1_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n1_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 754 17 34 10
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 3 0 1 0
## SIRA 396 156 489 306 561 573 780
##
## Overall Statistics
##
## Accuracy : 0.3762
## 95% CI : (0.3613, 0.3913)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2134
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.7093
## Specificity 1.00000 1.00000 1.0000 0.9798
## Pos Pred Value NaN NaN NaN 0.9252
## Neg Pred Value 0.90294 0.96176 0.8801 0.9054
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.1848
## Detection Prevalence 0.00000 0.00000 0.0000 0.1998
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8445
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.0016447 0.9873
## Specificity 1.0000 0.9991359 0.2459
## Pos Pred Value NaN 0.2500000 0.2392
## Neg Pred Value 0.8583 0.8510795 0.9878
## Prevalence 0.1417 0.1490196 0.1936
## Detection Rate 0.0000 0.0002451 0.1912
## Detection Prevalence 0.0000 0.0009804 0.7993
## Balanced Accuracy 0.5000 0.5003903 0.6166
db_tda_pc_5.50.5_n1_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 754 17 34 10
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 3 0 1 0
## SIRA 396 156 489 306 561 573 780
##
## Overall Statistics
##
## Accuracy : 0.3762
## 95% CI : (0.3613, 0.3913)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2134
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.7093
## Specificity 1.00000 1.00000 1.0000 0.9798
## Pos Pred Value NaN NaN NaN 0.9252
## Neg Pred Value 0.90294 0.96176 0.8801 0.9054
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.1848
## Detection Prevalence 0.00000 0.00000 0.0000 0.1998
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8445
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.0016447 0.9873
## Specificity 1.0000 0.9991359 0.2459
## Pos Pred Value NaN 0.2500000 0.2392
## Neg Pred Value 0.8583 0.8510795 0.9878
## Prevalence 0.1417 0.1490196 0.1936
## Detection Rate 0.0000 0.0002451 0.1912
## Detection Prevalence 0.0000 0.0009804 0.7993
## Balanced Accuracy 0.5000 0.5003903 0.6166
db_tda_pc_5.50.5_n1_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.762255e-01 2.134490e-01 3.613313e-01 3.912971e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 3.855318e-59 NaN
db_tda_pc_5.50.5_n1_db_nn1_cf0_ov_acc<-db_tda_pc_5.50.5_n1_db_nn1_cf0$overall[1]
db_tda_pc_5.50.5_n1_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.000000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.000000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.000000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.709313264 0.9797812 0.9251534 0.9053599 0.9251534
## Class: HOROZ 0.000000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.001644737 0.9991359 0.2500000 0.8510795 0.2500000
## Class: SIRA 0.987341772 0.2458967 0.2391904 0.9877900 0.2391904
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.000000000 NA 0.09705882 0.000000000
## Class: BOMBAY 0.000000000 NA 0.03823529 0.000000000
## Class: CALI 0.000000000 NA 0.11985294 0.000000000
## Class: DERMASON 0.709313264 0.802981896 0.26053922 0.184803922
## Class: HOROZ 0.000000000 NA 0.14166667 0.000000000
## Class: SEKER 0.001644737 0.003267974 0.14901961 0.000245098
## Class: SIRA 0.987341772 0.385090101 0.19362745 0.191176471
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000000 0.5000000
## Class: BOMBAY 0.0000000000 0.5000000
## Class: CALI 0.0000000000 0.5000000
## Class: DERMASON 0.1997549020 0.8445473
## Class: HOROZ 0.0000000000 0.5000000
## Class: SEKER 0.0009803922 0.5003903
## Class: SIRA 0.7992647059 0.6166192
db_tda_pc_5.50.5_n1_db_nn1_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n1_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold<-(db_nn1_fit_re - db_tda_pc_5.50.5_n1_nn1_fit_re)
diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold
## Accuracy
## 1 -0.1841773
## 2 -0.2782648
## 3 -0.3246575
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold
## $winLeft
## [1] 0.9911
##
## $winRope
## [1] 0.0089
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n1_3_fold
## $left
## [1] 0.983029
##
## $rope
## [1] 0.002296291
##
## $right
## [1] 0.01467474
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold))
#bf_tda_pca_5.50.5_nn1.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n1_3_fold)
## t = -6.3489, df = 2, p-value = 0.02392
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.44017309 -0.08455999
## sample estimates:
## mean of x
## -0.2623665
### Test set diff
diff_drybean_tda_pca_5.50.5_nn1.n1_test<-(db_nn1_cf_ov_acc - db_tda_pc_5.50.5_n1_db_nn1_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nn1.n1_test
## Accuracy
## 0.2335784
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n1_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n1_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n1_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n1_test$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n1_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1570333
##
## $winRight
## [1] 0.8429667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n1_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nn1.n1_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n1_test)) #bf_tda_pca_5.50.5_nn1.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n1_test))
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node2
##DryBean_TDA_PC_5.50.5_n2_NN1Fit0 <- nnet(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec, size=2, range = 0.6,, type='class')
#Neural Network 1
DryBean_TDA_PC_5.50.5_n2_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 55
## initial value 12736.191538
## iter 10 value 10610.958280
## iter 20 value 10603.664209
## iter 30 value 10594.123255
## iter 40 value 9026.794862
## iter 50 value 8543.245991
## iter 60 value 8286.361469
## iter 70 value 8130.579930
## iter 80 value 8045.208899
## iter 90 value 8003.465701
## iter 100 value 7869.928072
## final value 7869.928072
## stopped after 100 iterations
## # weights: 79
## initial value 15223.202922
## iter 10 value 10710.258688
## iter 20 value 10605.042773
## iter 30 value 10476.472891
## iter 40 value 8753.212144
## iter 50 value 8047.567014
## iter 60 value 7908.533163
## iter 70 value 7801.096546
## iter 80 value 7587.094129
## iter 90 value 7404.869795
## iter 100 value 7333.645771
## final value 7333.645771
## stopped after 100 iterations
## # weights: 127
## initial value 17975.538196
## iter 10 value 10602.393681
## iter 20 value 10598.065706
## iter 30 value 10597.323422
## iter 40 value 10597.228972
## iter 50 value 10597.138297
## final value 10597.137996
## converged
## # weights: 175
## initial value 14075.783447
## iter 10 value 10612.085956
## iter 20 value 10604.115645
## iter 30 value 10597.897086
## iter 40 value 9139.633220
## iter 50 value 8795.782568
## iter 60 value 8559.120816
## iter 70 value 8240.052884
## iter 80 value 7186.620864
## iter 90 value 6726.283999
## iter 100 value 6462.181534
## final value 6462.181534
## stopped after 100 iterations
## # weights: 55
## initial value 14698.319479
## iter 10 value 10924.945070
## iter 20 value 10644.121566
## iter 30 value 10603.955040
## iter 40 value 10603.146437
## final value 10603.143155
## converged
## # weights: 79
## initial value 12488.627179
## iter 10 value 10705.099613
## iter 20 value 10611.796473
## iter 30 value 10601.303556
## iter 40 value 10600.676717
## final value 10600.660746
## converged
## # weights: 127
## initial value 15086.492630
## iter 10 value 10647.902160
## iter 20 value 10603.203200
## iter 30 value 10366.728790
## iter 40 value 8419.139641
## iter 50 value 7703.566279
## iter 60 value 7504.043825
## iter 70 value 7152.511085
## iter 80 value 6898.527692
## iter 90 value 6514.681298
## iter 100 value 6331.364532
## final value 6331.364532
## stopped after 100 iterations
## # weights: 175
## initial value 14859.020836
## iter 10 value 10632.374701
## iter 20 value 9153.199996
## iter 30 value 8542.551332
## iter 40 value 8500.232819
## iter 50 value 7702.286424
## iter 60 value 6301.716804
## iter 70 value 6161.730002
## iter 80 value 5918.295249
## iter 90 value 5577.952333
## iter 100 value 5041.921312
## final value 5041.921312
## stopped after 100 iterations
## # weights: 55
## initial value 13923.153351
## iter 10 value 10668.275834
## iter 20 value 10605.907297
## final value 10605.901142
## converged
## # weights: 79
## initial value 14147.593684
## iter 10 value 10643.817657
## final value 10602.661509
## converged
## # weights: 127
## initial value 14315.155410
## iter 10 value 10662.049478
## iter 20 value 10602.478466
## iter 30 value 10431.233024
## iter 40 value 8346.444445
## iter 50 value 8214.412817
## iter 60 value 7885.296122
## iter 70 value 7122.146826
## iter 80 value 6734.549797
## iter 90 value 6074.458837
## iter 100 value 5854.471652
## final value 5854.471652
## stopped after 100 iterations
## # weights: 175
## initial value 18164.868195
## iter 10 value 10730.845897
## iter 20 value 10608.253897
## iter 30 value 10599.486946
## iter 40 value 10579.713690
## iter 50 value 9471.201045
## iter 60 value 8524.548241
## iter 70 value 7936.473149
## iter 80 value 7857.725430
## iter 90 value 7801.447906
## iter 100 value 7322.140276
## final value 7322.140276
## stopped after 100 iterations
## # weights: 55
## initial value 14425.417301
## iter 10 value 10754.426854
## iter 20 value 10603.068720
## iter 30 value 10469.581909
## iter 40 value 9937.170262
## iter 50 value 8115.574635
## iter 60 value 7914.761603
## iter 70 value 7270.327200
## iter 80 value 6291.700198
## iter 90 value 5151.550613
## iter 100 value 4046.431264
## final value 4046.431264
## stopped after 100 iterations
## # weights: 79
## initial value 14011.124602
## iter 10 value 10639.958485
## iter 20 value 10612.051107
## iter 30 value 10599.426996
## iter 40 value 10599.332461
## iter 50 value 10599.196640
## final value 10599.184427
## converged
## # weights: 127
## initial value 15100.699472
## iter 10 value 10619.946660
## iter 20 value 10605.178379
## iter 30 value 10599.380075
## iter 40 value 10597.424856
## iter 50 value 8201.096442
## iter 60 value 8132.392657
## iter 70 value 7991.841543
## iter 80 value 7601.986692
## iter 90 value 6519.010766
## iter 100 value 5171.414284
## final value 5171.414284
## stopped after 100 iterations
## # weights: 175
## initial value 14168.668569
## iter 10 value 10964.287442
## iter 20 value 10611.986849
## iter 30 value 10546.377277
## iter 40 value 10506.015065
## iter 50 value 10467.759046
## iter 60 value 10236.744915
## iter 70 value 8912.945840
## iter 80 value 6360.964035
## iter 90 value 6087.883597
## iter 100 value 5954.975783
## final value 5954.975783
## stopped after 100 iterations
## # weights: 55
## initial value 12308.107549
## iter 10 value 10690.081465
## iter 20 value 10609.275796
## iter 30 value 10602.215175
## final value 10602.199729
## converged
## # weights: 79
## initial value 14930.161860
## iter 10 value 10720.077225
## iter 20 value 10617.375102
## iter 30 value 10604.757070
## iter 40 value 10503.750333
## iter 50 value 9434.723210
## iter 60 value 9123.408338
## iter 70 value 8490.301283
## iter 80 value 7579.851671
## iter 90 value 7105.289410
## iter 100 value 6624.444419
## final value 6624.444419
## stopped after 100 iterations
## # weights: 127
## initial value 15085.188121
## iter 10 value 10680.810785
## iter 20 value 10610.547047
## iter 30 value 10602.006149
## iter 40 value 10601.843221
## iter 50 value 10601.143243
## iter 60 value 10600.772776
## iter 70 value 10595.179204
## iter 80 value 9297.320251
## iter 90 value 8159.967856
## iter 100 value 7923.121024
## final value 7923.121024
## stopped after 100 iterations
## # weights: 175
## initial value 17107.846923
## iter 10 value 10610.602014
## iter 20 value 10603.752383
## iter 30 value 10600.923381
## iter 40 value 10599.078948
## iter 50 value 10598.175340
## iter 60 value 10289.616507
## iter 70 value 8426.437341
## iter 80 value 8167.018635
## iter 90 value 8112.926579
## iter 100 value 8097.898047
## final value 8097.898047
## stopped after 100 iterations
## # weights: 55
## initial value 14469.609975
## iter 10 value 10637.957389
## iter 20 value 10604.263707
## final value 10604.208496
## converged
## # weights: 79
## initial value 12845.036227
## iter 10 value 10672.422383
## iter 20 value 10613.507709
## iter 30 value 10602.683348
## iter 40 value 10602.606587
## iter 50 value 10602.460917
## final value 10602.457321
## converged
## # weights: 127
## initial value 13106.767814
## iter 10 value 10668.221736
## iter 20 value 10615.791172
## iter 30 value 10602.947094
## iter 40 value 10601.763132
## iter 50 value 10579.371388
## iter 60 value 9069.789626
## iter 70 value 8983.033983
## iter 80 value 6790.773906
## iter 90 value 6227.468044
## iter 100 value 5791.039611
## final value 5791.039611
## stopped after 100 iterations
## # weights: 175
## initial value 13996.973073
## iter 10 value 10604.148276
## iter 20 value 10600.647894
## iter 30 value 10600.603039
## iter 40 value 10600.598263
## iter 40 value 10600.598158
## iter 50 value 10600.191994
## iter 60 value 8700.121422
## iter 70 value 8562.131718
## iter 80 value 8338.511242
## iter 90 value 8222.542915
## iter 100 value 7252.649278
## final value 7252.649278
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 52
## initial value 11189.484925
## iter 10 value 10581.174010
## iter 10 value 10581.173960
## iter 10 value 10581.173906
## final value 10581.173906
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 75
## initial value 13588.630109
## iter 10 value 10581.916132
## iter 20 value 10581.182581
## iter 30 value 10581.074803
## iter 40 value 8658.686645
## iter 50 value 8024.496778
## iter 60 value 7933.140620
## iter 70 value 7423.657824
## iter 80 value 6256.298340
## iter 90 value 5689.819030
## iter 100 value 5308.115401
## final value 5308.115401
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 121
## initial value 13453.771182
## iter 10 value 10581.152121
## final value 10581.041595
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 167
## initial value 11645.128186
## iter 10 value 10581.042469
## iter 20 value 9837.678711
## iter 30 value 8885.168951
## iter 40 value 8103.479241
## iter 50 value 7518.083637
## iter 60 value 7425.997854
## iter 70 value 7173.568794
## iter 80 value 6770.507175
## iter 90 value 6579.863858
## iter 100 value 6460.683520
## final value 6460.683520
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 52
## initial value 11629.586881
## iter 10 value 10582.435588
## iter 20 value 10581.471911
## iter 30 value 10581.437337
## iter 40 value 10581.336906
## iter 50 value 10581.216179
## final value 10581.206452
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 75
## initial value 13205.845075
## iter 10 value 10582.953110
## iter 20 value 10581.223361
## iter 30 value 10525.093344
## iter 40 value 8737.007103
## iter 50 value 8167.696441
## iter 60 value 7278.096954
## iter 70 value 6565.414790
## iter 80 value 6094.660487
## iter 90 value 5213.477640
## iter 100 value 5005.826073
## final value 5005.826073
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 121
## initial value 13415.109462
## iter 10 value 10591.527472
## iter 20 value 10581.174091
## iter 30 value 10581.153828
## iter 30 value 10581.153779
## iter 40 value 10581.060825
## final value 10581.056416
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 167
## initial value 11967.304456
## iter 10 value 10602.591394
## iter 20 value 10581.330890
## iter 30 value 10581.088468
## iter 40 value 10533.825874
## iter 50 value 10497.001765
## iter 60 value 9532.608338
## iter 70 value 8437.647957
## iter 80 value 8324.462396
## iter 90 value 8321.409880
## iter 100 value 8295.244109
## final value 8295.244109
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 52
## initial value 11316.816399
## iter 10 value 10587.008746
## iter 20 value 10582.299848
## iter 30 value 10580.837620
## iter 40 value 9703.592902
## iter 50 value 8605.834346
## iter 60 value 8245.387152
## iter 70 value 8099.623639
## iter 80 value 7960.785012
## iter 90 value 7924.541603
## iter 100 value 7883.177412
## final value 7883.177412
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 75
## initial value 11748.898205
## iter 10 value 10581.527250
## iter 20 value 10581.369730
## iter 30 value 8710.491725
## iter 40 value 8340.265604
## iter 50 value 7928.199532
## iter 60 value 7130.627731
## iter 70 value 6683.310983
## iter 80 value 6346.060428
## iter 90 value 5796.672058
## iter 100 value 5502.126886
## final value 5502.126886
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 121
## initial value 10804.338800
## iter 10 value 10581.158227
## final value 10581.156325
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'BOMBAY' is empty
## # weights: 167
## initial value 11902.177813
## iter 10 value 10581.547693
## iter 20 value 10581.516516
## iter 20 value 10581.516476
## iter 30 value 10492.412951
## iter 40 value 10128.484116
## iter 50 value 9632.090284
## iter 60 value 8688.194705
## iter 70 value 7958.138650
## iter 80 value 6707.604316
## iter 90 value 6447.330221
## iter 100 value 6417.920097
## final value 6417.920097
## stopped after 100 iterations
## # weights: 175
## initial value 25196.346574
## iter 10 value 16163.956115
## iter 20 value 15946.809606
## iter 30 value 15901.223883
## iter 40 value 15889.681445
## iter 50 value 15889.665570
## iter 60 value 15888.704245
## final value 15888.606791
## converged
DryBean_TDA_PC_5.50.5_n2_NN1Fit0
## Neural Network
##
## 9515 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6344, 6345, 6341
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.4301061 0.32129608
## 2 0.5 0.2044674 0.00000000
## 2 0.7 0.2422746 0.03542725
## 3 0.3 0.2949376 0.10782251
## 3 0.5 0.3369107 0.16718787
## 3 0.7 0.1898696 -0.04066983
## 5 0.3 0.3543097 0.21592629
## 5 0.5 0.3686806 0.25256830
## 5 0.7 0.4161126 0.32071143
## 7 0.3 0.4683291 0.34835435
## 7 0.5 0.4051630 0.28740099
## 7 0.7 0.4004130 0.28022583
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 7 and decay = 0.3.
DryBean_TDA_PC_5.50.5_n2_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.6602524 0.560514774 Fold2
## 2 0.5963418 0.481550296 Fold1
## 3 0.1483932 0.002997983 Fold3
db_tda_pc_5.50.5_n2_nn1_fit_re<-DryBean_TDA_PC_5.50.5_n2_NN1Fit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n2_NN1Fit0)
## a 16-7-7 network with 175 weights
## options were - softmax modelling decay=0.3
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h6 i1->h6 i2->h6 i3->h6 i4->h6 i5->h6 i6->h6 i7->h6 i8->h6 i9->h6
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h6 i11->h6 i12->h6 i13->h6 i14->h6 i15->h6 i16->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h7 i1->h7 i2->h7 i3->h7 i4->h7 i5->h7 i6->h7 i7->h7 i8->h7 i9->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h7 i11->h7 i12->h7 i13->h7 i14->h7 i15->h7 i16->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1 h6->o1 h7->o1
## 0.14 0.00 0.14 0.14 0.14 0.14 0.14 0.00
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2 h6->o2 h7->o2
## -0.97 0.01 -0.97 -0.97 -0.97 -0.97 -0.97 0.00
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3 h6->o3 h7->o3
## 0.10 0.00 0.10 0.10 0.10 0.10 0.10 0.00
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4 h6->o4 h7->o4
## 0.26 -0.01 0.26 0.26 0.26 0.26 0.26 0.00
## b->o5 h1->o5 h2->o5 h3->o5 h4->o5 h5->o5 h6->o5 h7->o5
## 0.18 0.00 0.18 0.18 0.19 0.18 0.18 0.00
## b->o6 h1->o6 h2->o6 h3->o6 h4->o6 h5->o6 h6->o6 h7->o6
## 0.01 0.00 0.01 0.01 0.02 0.01 0.01 0.00
## b->o7 h1->o7 h2->o7 h3->o7 h4->o7 h5->o7 h6->o7 h7->o7
## 0.27 -0.01 0.27 0.27 0.27 0.27 0.27 0.00
#vip(DryBean_TDA_PC_5.50.5_n2_NN1Fit0,25) + ggtitle("dryBean_TDA_PCA_5.50.5_n2_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_PC_5.50.5_n2_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n2_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n2_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n2_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 396 156 489 1063 578 608 790
##
## Overall Statistics
##
## Accuracy : 0.1936
## 95% CI : (0.1816, 0.2061)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.0000
## Specificity 1.00000 1.00000 1.0000 1.0000
## Pos Pred Value NaN NaN NaN NaN
## Neg Pred Value 0.90294 0.96176 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.0000
## Detection Prevalence 0.00000 0.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 1.0000
## Specificity 1.0000 1.000 0.0000
## Pos Pred Value NaN NaN 0.1936
## Neg Pred Value 0.8583 0.851 NaN
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.1936
## Detection Prevalence 0.0000 0.000 1.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n2_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 396 156 489 1063 578 608 790
##
## Overall Statistics
##
## Accuracy : 0.1936
## 95% CI : (0.1816, 0.2061)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.0000
## Specificity 1.00000 1.00000 1.0000 1.0000
## Pos Pred Value NaN NaN NaN NaN
## Neg Pred Value 0.90294 0.96176 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.0000
## Detection Prevalence 0.00000 0.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 1.0000
## Specificity 1.0000 1.000 0.0000
## Pos Pred Value NaN NaN 0.1936
## Neg Pred Value 0.8583 0.851 NaN
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.1936
## Detection Prevalence 0.0000 0.000 1.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n2_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.1936275 0.0000000 0.1816029 0.2060923 0.2605392
## AccuracyPValue McnemarPValue
## 1.0000000 NaN
db_tda_pc_5.50.5_n2_db_nn1_cf0_ov_acc<-db_tda_pc_5.50.5_n2_db_nn1_cf0$overall[1]
db_tda_pc_5.50.5_n2_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0 1 NaN 0.9029412 NA
## Class: BOMBAY 0 1 NaN 0.9617647 NA
## Class: CALI 0 1 NaN 0.8801471 NA
## Class: DERMASON 0 1 NaN 0.7394608 NA
## Class: HOROZ 0 1 NaN 0.8583333 NA
## Class: SEKER 0 1 NaN 0.8509804 NA
## Class: SIRA 1 0 0.1936275 NaN 0.1936275
## Recall F1 Prevalence Detection Rate Detection Prevalence
## Class: BARBUNYA 0 NA 0.09705882 0.0000000 0
## Class: BOMBAY 0 NA 0.03823529 0.0000000 0
## Class: CALI 0 NA 0.11985294 0.0000000 0
## Class: DERMASON 0 NA 0.26053922 0.0000000 0
## Class: HOROZ 0 NA 0.14166667 0.0000000 0
## Class: SEKER 0 NA 0.14901961 0.0000000 0
## Class: SIRA 1 0.3244353 0.19362745 0.1936275 1
## Balanced Accuracy
## Class: BARBUNYA 0.5
## Class: BOMBAY 0.5
## Class: CALI 0.5
## Class: DERMASON 0.5
## Class: HOROZ 0.5
## Class: SEKER 0.5
## Class: SIRA 0.5
db_tda_pc_5.50.5_n2_db_nn1_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n2_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold<-(db_nn1_fit_re - db_tda_pc_5.50.5_n2_nn1_fit_re)
diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold
## Accuracy
## 1 0.04535217
## 2 -0.06313442
## 3 0.41894475
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold
## $probLeft
## [1] 0.25
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold_odds.left
## [1] 0.5
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold
## $winLeft
## [1] 0.1529333
##
## $winRope
## [1] 0.046
##
## $winRight
## [1] 0.8010667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n2_3_fold
## $left
## [1] 0.2418801
##
## $rope
## [1] 0.02783128
##
## $right
## [1] 0.7302886
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold))
#bf_tda_pca_5.50.5_nn1.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n2_3_fold)
## t = 0.91583, df = 2, p-value = 0.4564
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.4945099 0.7619516
## sample estimates:
## mean of x
## 0.1337208
### Test set diff
diff_drybean_tda_pca_5.50.5_nn1.n2_test<-(db_nn1_cf_ov_acc - db_tda_pc_5.50.5_n2_db_nn1_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nn1.n2_test
## Accuracy
## 0.4161765
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n2_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n2_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n2_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n2_test$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n2_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1598667
##
## $winRight
## [1] 0.8401333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n2_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nn1.n2_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n2_test)) #bf_tda_pca_5.50.5_nn1.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n2_test))
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node3
#Neural Network 1
DryBean_TDA_PC_5.50.5_n3_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 55
## initial value 6722.184776
## iter 10 value 4880.085496
## iter 20 value 4864.168943
## iter 30 value 4820.612062
## iter 40 value 3711.199166
## iter 50 value 3289.843726
## iter 60 value 3214.266146
## iter 70 value 3183.495580
## iter 80 value 3144.069776
## iter 90 value 3111.158053
## iter 100 value 2825.772144
## final value 2825.772144
## stopped after 100 iterations
## # weights: 79
## initial value 6716.838583
## iter 10 value 4868.725792
## iter 20 value 4862.207902
## iter 30 value 4821.430972
## iter 40 value 4504.013033
## iter 50 value 4176.593648
## iter 60 value 3498.853614
## iter 70 value 2944.729837
## iter 80 value 2273.817040
## iter 90 value 2189.644863
## iter 100 value 2122.372451
## final value 2122.372451
## stopped after 100 iterations
## # weights: 127
## initial value 7613.496328
## iter 10 value 4914.879776
## iter 20 value 4872.932557
## iter 30 value 4857.735167
## iter 40 value 4853.071814
## iter 50 value 4816.116692
## iter 60 value 4793.758317
## iter 70 value 3286.240689
## iter 80 value 2699.811847
## iter 90 value 2681.938835
## iter 100 value 2437.476463
## final value 2437.476463
## stopped after 100 iterations
## # weights: 175
## initial value 8319.362306
## iter 10 value 4887.897632
## iter 20 value 4863.614173
## iter 30 value 4835.620148
## iter 40 value 4549.832362
## iter 50 value 3672.862397
## iter 60 value 3293.454696
## iter 70 value 3149.939440
## iter 80 value 2847.062378
## iter 90 value 2540.411951
## iter 100 value 2356.711175
## final value 2356.711175
## stopped after 100 iterations
## # weights: 55
## initial value 6924.851133
## iter 10 value 4952.512991
## iter 20 value 4872.642039
## iter 30 value 4864.292511
## iter 40 value 4861.717926
## iter 50 value 4816.470105
## iter 60 value 4748.741424
## iter 70 value 3727.750457
## iter 80 value 3514.077902
## iter 90 value 3172.501580
## iter 100 value 2962.225516
## final value 2962.225516
## stopped after 100 iterations
## # weights: 79
## initial value 6277.988583
## iter 10 value 4894.206976
## iter 20 value 4877.737113
## iter 30 value 4842.409833
## iter 40 value 4581.382538
## iter 50 value 3931.128364
## iter 60 value 3739.517081
## iter 70 value 3312.755480
## iter 80 value 2722.083974
## iter 90 value 2419.110802
## iter 100 value 2234.926909
## final value 2234.926909
## stopped after 100 iterations
## # weights: 127
## initial value 7033.837268
## iter 10 value 4883.930115
## iter 20 value 4867.014020
## iter 30 value 4689.979608
## iter 40 value 4470.136068
## iter 50 value 4097.878016
## iter 60 value 3467.013629
## iter 70 value 3253.177569
## iter 80 value 3040.282242
## iter 90 value 2685.106476
## iter 100 value 2492.530175
## final value 2492.530175
## stopped after 100 iterations
## # weights: 175
## initial value 7911.488260
## iter 10 value 5266.787169
## iter 20 value 4958.521698
## iter 30 value 4857.748285
## iter 40 value 4454.069462
## iter 50 value 3351.050146
## iter 60 value 3099.489960
## iter 70 value 2940.854461
## iter 80 value 2707.588683
## iter 90 value 2679.650324
## iter 100 value 2621.775739
## final value 2621.775739
## stopped after 100 iterations
## # weights: 55
## initial value 7069.548851
## iter 10 value 5023.577505
## iter 20 value 4895.666388
## iter 30 value 4884.368010
## iter 40 value 4883.960732
## iter 50 value 4883.908047
## iter 60 value 4866.495837
## iter 70 value 3741.589032
## iter 80 value 3573.564991
## iter 90 value 3319.162514
## iter 100 value 3120.743704
## final value 3120.743704
## stopped after 100 iterations
## # weights: 79
## initial value 7361.088884
## iter 10 value 4968.133912
## iter 20 value 4879.953178
## iter 30 value 4864.812446
## iter 40 value 4859.941803
## iter 50 value 4279.473637
## iter 60 value 3895.750342
## iter 70 value 3598.139677
## iter 80 value 3386.481672
## iter 90 value 2644.524899
## iter 100 value 2433.383792
## final value 2433.383792
## stopped after 100 iterations
## # weights: 127
## initial value 6624.404102
## iter 10 value 4936.745512
## iter 20 value 4872.995873
## iter 30 value 4549.916082
## iter 40 value 4104.852438
## iter 50 value 3445.176517
## iter 60 value 2781.552432
## iter 70 value 2529.018984
## iter 80 value 2487.946434
## iter 90 value 2449.850524
## iter 100 value 2230.837063
## final value 2230.837063
## stopped after 100 iterations
## # weights: 175
## initial value 10191.401842
## iter 10 value 4919.607541
## iter 20 value 4806.435993
## iter 30 value 3673.996280
## iter 40 value 3633.991763
## iter 50 value 3346.379626
## iter 60 value 2669.298381
## iter 70 value 2553.856481
## iter 80 value 2540.628710
## iter 90 value 2531.975559
## iter 100 value 2492.519715
## final value 2492.519715
## stopped after 100 iterations
## # weights: 55
## initial value 6872.855300
## iter 10 value 4971.285214
## iter 20 value 4871.760966
## iter 30 value 4799.614266
## iter 40 value 4587.213291
## iter 50 value 4289.193849
## iter 60 value 3897.765471
## iter 70 value 3290.407648
## iter 80 value 2864.108798
## iter 90 value 2722.930601
## iter 100 value 2659.966331
## final value 2659.966331
## stopped after 100 iterations
## # weights: 79
## initial value 7104.684945
## iter 10 value 4870.789446
## iter 20 value 4868.893892
## iter 30 value 4626.886827
## iter 40 value 4286.935098
## iter 50 value 3829.002347
## iter 60 value 3091.330690
## iter 70 value 2970.467683
## iter 80 value 2782.485949
## iter 90 value 2404.158297
## iter 100 value 2111.710496
## final value 2111.710496
## stopped after 100 iterations
## # weights: 127
## initial value 7653.600939
## iter 10 value 4894.447787
## iter 20 value 4869.335920
## iter 30 value 4865.206875
## iter 40 value 4864.432872
## iter 50 value 4863.842286
## iter 60 value 4236.542810
## iter 70 value 4104.278662
## iter 80 value 3831.291856
## iter 90 value 3598.097464
## iter 100 value 3498.525453
## final value 3498.525453
## stopped after 100 iterations
## # weights: 175
## initial value 8495.352048
## iter 10 value 4960.810730
## iter 20 value 4873.774066
## iter 30 value 4869.660613
## iter 40 value 4107.011162
## iter 50 value 3480.275009
## iter 60 value 3347.434724
## iter 70 value 3256.316366
## iter 80 value 3052.015448
## iter 90 value 2985.581281
## iter 100 value 2913.522040
## final value 2913.522040
## stopped after 100 iterations
## # weights: 55
## initial value 7929.646289
## iter 10 value 4889.583282
## iter 20 value 4873.265291
## iter 30 value 4833.973414
## iter 40 value 4311.959087
## iter 50 value 3387.806144
## iter 60 value 2638.432141
## iter 70 value 2296.368105
## iter 80 value 2234.966812
## iter 90 value 2203.600098
## iter 100 value 2198.865484
## final value 2198.865484
## stopped after 100 iterations
## # weights: 79
## initial value 7639.107229
## iter 10 value 4878.923419
## iter 20 value 4874.733280
## iter 30 value 4193.317105
## iter 40 value 3745.557729
## iter 50 value 3225.703409
## iter 60 value 3165.218959
## iter 70 value 3154.072650
## iter 80 value 3067.354891
## iter 90 value 3017.689349
## iter 100 value 2902.639352
## final value 2902.639352
## stopped after 100 iterations
## # weights: 127
## initial value 6246.533951
## iter 10 value 4885.511667
## iter 20 value 4868.434272
## iter 30 value 4867.322548
## iter 40 value 4865.953825
## iter 50 value 3397.449430
## iter 60 value 2594.975318
## iter 70 value 2454.290663
## iter 80 value 2331.295090
## iter 90 value 2066.900776
## iter 100 value 1848.499644
## final value 1848.499644
## stopped after 100 iterations
## # weights: 175
## initial value 8371.015412
## iter 10 value 4973.883408
## iter 20 value 4880.599278
## iter 30 value 4876.903572
## iter 40 value 4875.883785
## iter 50 value 4853.518878
## iter 60 value 4577.092601
## iter 70 value 4521.623501
## iter 80 value 4437.556803
## iter 90 value 4372.400446
## iter 100 value 4078.776473
## final value 4078.776473
## stopped after 100 iterations
## # weights: 55
## initial value 6944.237666
## iter 10 value 4889.293286
## iter 20 value 4874.775707
## iter 30 value 4869.487860
## iter 40 value 4867.640468
## iter 50 value 4867.377780
## iter 60 value 4532.237764
## iter 70 value 3872.712453
## iter 80 value 3670.475708
## iter 90 value 3379.926343
## iter 100 value 3089.101117
## final value 3089.101117
## stopped after 100 iterations
## # weights: 79
## initial value 7780.948790
## iter 10 value 4885.740287
## iter 20 value 4877.297495
## iter 30 value 3400.374926
## iter 40 value 2995.665489
## iter 50 value 2864.364162
## iter 60 value 2493.250596
## iter 70 value 2295.694386
## iter 80 value 2160.966139
## iter 90 value 1943.938148
## iter 100 value 1728.400159
## final value 1728.400159
## stopped after 100 iterations
## # weights: 127
## initial value 7643.904938
## iter 10 value 4944.152377
## iter 20 value 4920.938228
## iter 30 value 4912.436515
## iter 40 value 3995.696258
## iter 50 value 3791.611134
## iter 60 value 3731.833309
## iter 70 value 3721.292403
## iter 80 value 3701.340945
## iter 90 value 2976.324524
## iter 100 value 2363.253150
## final value 2363.253150
## stopped after 100 iterations
## # weights: 175
## initial value 6810.211284
## iter 10 value 4903.172697
## iter 20 value 4870.474604
## iter 30 value 4867.293403
## iter 40 value 4771.266401
## iter 50 value 4685.482735
## iter 60 value 3811.685130
## iter 70 value 3198.771741
## iter 80 value 2759.823472
## iter 90 value 2417.944804
## iter 100 value 2352.905572
## final value 2352.905572
## stopped after 100 iterations
## # weights: 55
## initial value 6990.397583
## iter 10 value 4898.041724
## iter 20 value 4865.568488
## iter 30 value 4864.171139
## iter 40 value 3535.493099
## iter 50 value 3236.991133
## iter 60 value 3056.723080
## iter 70 value 2949.213652
## iter 80 value 2904.297700
## iter 90 value 2833.328505
## iter 100 value 2813.550727
## final value 2813.550727
## stopped after 100 iterations
## # weights: 79
## initial value 10108.106129
## iter 10 value 4868.523457
## iter 20 value 4868.211361
## iter 30 value 4782.171557
## iter 40 value 4470.542694
## iter 50 value 3546.052780
## iter 60 value 2467.906855
## iter 70 value 1983.581494
## iter 80 value 1800.802476
## iter 90 value 1230.571966
## iter 100 value 982.099720
## final value 982.099720
## stopped after 100 iterations
## # weights: 127
## initial value 8007.100939
## iter 10 value 4941.614892
## iter 20 value 4867.700407
## iter 30 value 4858.668404
## iter 40 value 3559.954505
## iter 50 value 3274.611795
## iter 60 value 3162.820968
## iter 70 value 3094.590076
## iter 80 value 2962.692734
## iter 90 value 2760.779170
## iter 100 value 1980.967320
## final value 1980.967320
## stopped after 100 iterations
## # weights: 175
## initial value 10400.892119
## iter 10 value 4882.899670
## iter 20 value 4872.749279
## iter 30 value 4860.937479
## iter 40 value 4658.323956
## iter 50 value 3343.644161
## iter 60 value 2683.897000
## iter 70 value 2088.557134
## iter 80 value 1807.565505
## iter 90 value 1674.711474
## iter 100 value 1600.459344
## final value 1600.459344
## stopped after 100 iterations
## # weights: 55
## initial value 7146.779859
## iter 10 value 4882.222525
## iter 20 value 4867.716289
## iter 30 value 4867.453676
## iter 40 value 4864.126562
## iter 50 value 3774.305281
## iter 60 value 3140.019291
## iter 70 value 3111.966855
## iter 80 value 2985.357391
## iter 90 value 2908.167417
## iter 100 value 2898.455001
## final value 2898.455001
## stopped after 100 iterations
## # weights: 79
## initial value 6695.368768
## iter 10 value 4927.757340
## iter 20 value 4867.027980
## iter 30 value 3674.517072
## iter 40 value 3473.446900
## iter 50 value 3397.032403
## iter 60 value 3040.325660
## iter 70 value 2834.887657
## iter 80 value 2728.477108
## iter 90 value 2665.982385
## iter 100 value 2484.670710
## final value 2484.670710
## stopped after 100 iterations
## # weights: 127
## initial value 8976.823892
## iter 10 value 5056.350915
## iter 20 value 4884.288482
## iter 30 value 4864.762217
## iter 40 value 4860.622092
## iter 50 value 3817.410277
## iter 60 value 3468.697034
## iter 70 value 3439.478600
## iter 80 value 3174.386930
## iter 90 value 2439.993210
## iter 100 value 2067.535269
## final value 2067.535269
## stopped after 100 iterations
## # weights: 175
## initial value 9238.455739
## iter 10 value 4906.331768
## iter 20 value 4899.475269
## iter 30 value 4863.361323
## iter 40 value 4862.573773
## iter 50 value 3680.511717
## iter 60 value 3312.440623
## iter 70 value 3280.365848
## iter 80 value 3201.105025
## iter 90 value 3170.534451
## iter 100 value 3031.896870
## final value 3031.896870
## stopped after 100 iterations
## # weights: 55
## initial value 8119.150739
## iter 10 value 4904.921314
## iter 20 value 4881.243491
## iter 30 value 4878.290697
## iter 40 value 3671.478107
## iter 50 value 3397.464755
## iter 60 value 3362.000649
## iter 70 value 3264.128067
## iter 80 value 2659.302454
## iter 90 value 2242.990524
## iter 100 value 1834.438735
## final value 1834.438735
## stopped after 100 iterations
## # weights: 79
## initial value 7996.221053
## iter 10 value 4871.294392
## iter 20 value 4870.218512
## iter 30 value 4869.936682
## iter 40 value 4100.173098
## iter 50 value 3111.748279
## iter 60 value 2988.496427
## iter 70 value 2892.548402
## iter 80 value 2628.189098
## iter 90 value 2592.293585
## iter 100 value 2279.461852
## final value 2279.461852
## stopped after 100 iterations
## # weights: 127
## initial value 7567.230997
## iter 10 value 4961.472935
## iter 20 value 4861.939081
## iter 30 value 4861.059181
## iter 40 value 4860.505769
## iter 50 value 4628.744784
## iter 60 value 4428.023611
## iter 70 value 3406.755882
## iter 80 value 3236.132591
## iter 90 value 2931.776803
## iter 100 value 2906.430831
## final value 2906.430831
## stopped after 100 iterations
## # weights: 175
## initial value 8728.371180
## iter 10 value 4891.900960
## iter 20 value 4867.671105
## iter 30 value 4864.699468
## iter 40 value 4701.099634
## iter 50 value 4179.862300
## iter 60 value 3884.927597
## iter 70 value 3618.082496
## iter 80 value 3485.101297
## iter 90 value 3317.734232
## iter 100 value 3132.649767
## final value 3132.649767
## stopped after 100 iterations
## # weights: 79
## initial value 9527.464660
## iter 10 value 7354.808236
## iter 20 value 7302.381537
## iter 30 value 7129.535192
## iter 40 value 5143.308648
## iter 50 value 4679.977705
## iter 60 value 4657.636977
## iter 70 value 4536.169310
## iter 80 value 4266.516622
## iter 90 value 4189.323548
## iter 100 value 4157.348083
## final value 4157.348083
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n3_NN1Fit0
## Neural Network
##
## 5355 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 3569, 3571, 3570
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.6425770 0.4691813
## 2 0.5 0.7044234 0.5648260
## 2 0.7 0.7142842 0.5800084
## 3 0.3 0.8558380 0.7925276
## 3 0.5 0.7390937 0.6196016
## 3 0.7 0.7785532 0.6909881
## 5 0.3 0.7467416 0.6291674
## 5 0.5 0.8101019 0.7297981
## 5 0.7 0.7344639 0.6137419
## 7 0.3 0.7706630 0.6685634
## 7 0.5 0.5984667 0.3993136
## 7 0.7 0.6476224 0.5068622
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 3 and decay = 0.3.
DryBean_TDA_PC_5.50.5_n3_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.9266106 0.8971866 Fold3
## 2 0.8267937 0.7484719 Fold2
## 3 0.8141097 0.7319242 Fold1
db_tda_pc_5.50.5_n3_nn1_fit_re<-DryBean_TDA_PC_5.50.5_n3_NN1Fit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n3_NN1Fit0)
## a 16-3-7 network with 79 weights
## options were - softmax modelling decay=0.3
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## -0.02 -1.07 -0.33 -1.54 -1.00 -0.07 -0.01 1.07 -0.25 0.07
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## -0.02 -0.02 -0.01 0.00 0.00 0.00 -0.02
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.09 0.03 -0.28 -3.83 -0.03 0.09 0.06 -0.02 4.46 0.64
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.09 0.14 0.09 0.00 0.00 0.09 0.13
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1
## -0.46 1.21 3.07 -0.46
## b->o2 h1->o2 h2->o2 h3->o2
## -1.16 0.99 2.34 -1.16
## b->o3 h1->o3 h2->o3 h3->o3
## 0.35 -0.11 2.65 0.35
## b->o4 h1->o4 h2->o4 h3->o4
## -1.45 1.05 -1.65 -1.45
## b->o5 h1->o5 h2->o5 h3->o5
## 2.66 -1.35 -5.03 2.66
## b->o6 h1->o6 h2->o6 h3->o6
## -1.24 0.00 -1.07 -1.24
## b->o7 h1->o7 h2->o7 h3->o7
## 1.30 -1.79 -0.31 1.30
#vip(DryBean_TDA_PC_5.50.5_n3_NN1Fit0,25) + ggtitle("dryBean_TDA_PCA_5.50.5_n3_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_PC_5.50.5_n3_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n3_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n3_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n3_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 312 144 261 8 8 11 34
## BOMBAY 0 0 0 0 0 0 0
## CALI 79 12 216 776 10 597 686
## DERMASON 0 0 0 0 0 0 0
## HOROZ 5 0 12 273 560 0 70
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 6 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2667
## 95% CI : (0.2531, 0.2805)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.1909
##
## Kappa : 0.1662
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.78788 0.00000 0.44172 0.0000
## Specificity 0.87351 1.00000 0.39850 1.0000
## Pos Pred Value 0.40103 NaN 0.09091 NaN
## Neg Pred Value 0.97456 0.96176 0.83979 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.07647 0.00000 0.05294 0.0000
## Detection Prevalence 0.19069 0.00000 0.58235 0.0000
## Balanced Accuracy 0.83069 0.50000 0.42011 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9689 0.000 0.000000
## Specificity 0.8972 1.000 0.998176
## Pos Pred Value 0.6087 NaN 0.000000
## Neg Pred Value 0.9943 0.851 0.806087
## Prevalence 0.1417 0.149 0.193627
## Detection Rate 0.1373 0.000 0.000000
## Detection Prevalence 0.2255 0.000 0.001471
## Balanced Accuracy 0.9330 0.500 0.499088
db_tda_pc_5.50.5_n3_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 312 144 261 8 8 11 34
## BOMBAY 0 0 0 0 0 0 0
## CALI 79 12 216 776 10 597 686
## DERMASON 0 0 0 0 0 0 0
## HOROZ 5 0 12 273 560 0 70
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 6 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2667
## 95% CI : (0.2531, 0.2805)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.1909
##
## Kappa : 0.1662
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.78788 0.00000 0.44172 0.0000
## Specificity 0.87351 1.00000 0.39850 1.0000
## Pos Pred Value 0.40103 NaN 0.09091 NaN
## Neg Pred Value 0.97456 0.96176 0.83979 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.07647 0.00000 0.05294 0.0000
## Detection Prevalence 0.19069 0.00000 0.58235 0.0000
## Balanced Accuracy 0.83069 0.50000 0.42011 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9689 0.000 0.000000
## Specificity 0.8972 1.000 0.998176
## Pos Pred Value 0.6087 NaN 0.000000
## Neg Pred Value 0.9943 0.851 0.806087
## Prevalence 0.1417 0.149 0.193627
## Detection Rate 0.1373 0.000 0.000000
## Detection Prevalence 0.2255 0.000 0.001471
## Balanced Accuracy 0.9330 0.500 0.499088
db_tda_pc_5.50.5_n3_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.2666667 0.1661610 0.2531459 0.2805223 0.2605392
## AccuracyPValue McnemarPValue
## 0.1909134 NaN
db_tda_pc_5.50.5_n3_db_nn1_cf0_ov_acc<-db_tda_pc_5.50.5_n3_db_nn1_cf0$overall[1]
db_tda_pc_5.50.5_n3_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0.7878788 0.8735071 0.40102828 0.9745609
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647
## Class: CALI 0.4417178 0.3984962 0.09090909 0.8397887
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608
## Class: HOROZ 0.9688581 0.8972016 0.60869565 0.9943038
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804
## Class: SIRA 0.0000000 0.9981763 0.00000000 0.8060874
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.40102828 0.7878788 0.5315162 0.09705882 0.07647059
## Class: BOMBAY NA 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.09090909 0.4417178 0.1507853 0.11985294 0.05294118
## Class: DERMASON NA 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.60869565 0.9688581 0.7476636 0.14166667 0.13725490
## Class: SEKER NA 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.00000000 0.0000000 NaN 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.190686275 0.8306929
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.582352941 0.4201070
## Class: DERMASON 0.000000000 0.5000000
## Class: HOROZ 0.225490196 0.9330299
## Class: SEKER 0.000000000 0.5000000
## Class: SIRA 0.001470588 0.4990881
db_tda_pc_5.50.5_n3_db_nn1_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n3_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold<-(db_nn1_fit_re - db_tda_pc_5.50.5_n3_nn1_fit_re)
diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold
## Accuracy
## 1 -0.2210061
## 2 -0.2935863
## 3 -0.2467718
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold
## $winLeft
## [1] 0.9909333
##
## $winRope
## [1] 0.009066667
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n3_3_fold
## $left
## [1] 0.9950133
##
## $rope
## [1] 0.0007182117
##
## $right
## [1] 0.004268467
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold))
#bf_tda_pca_5.50.5_nn1.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n3_3_fold)
## t = -11.946, df = 2, p-value = 0.006934
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.3451926 -0.1623836
## sample estimates:
## mean of x
## -0.2537881
### Test set diff
diff_drybean_tda_pca_5.50.5_nn1.n3_test<-(db_nn1_cf_ov_acc - db_tda_pc_5.50.5_n3_db_nn1_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nn1.n3_test
## Accuracy
## 0.3431373
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n3_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n3_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n3_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n3_test$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n3_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1569667
##
## $winRight
## [1] 0.8430333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n3_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nn1.n3_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n3_test)) #bf_tda_pca_5.50.5_nn1.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n3_test))
##Node4
#Neural Network 1
DryBean_TDA_PC_5.50.5_n4_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 46
## initial value 1423.757504
## iter 10 value 1322.547303
## iter 20 value 1179.442279
## iter 30 value 953.664190
## iter 40 value 787.809077
## iter 50 value 748.254529
## iter 60 value 734.887202
## iter 70 value 723.017604
## iter 80 value 714.355382
## iter 90 value 712.940930
## iter 100 value 650.937639
## final value 650.937639
## stopped after 100 iterations
## # weights: 67
## initial value 1390.641654
## iter 10 value 1324.492368
## iter 20 value 871.764487
## iter 30 value 728.993154
## iter 40 value 724.025447
## iter 50 value 721.612294
## iter 60 value 662.305491
## iter 70 value 486.411475
## iter 80 value 407.993551
## iter 90 value 399.988362
## iter 100 value 375.549522
## final value 375.549522
## stopped after 100 iterations
## # weights: 109
## initial value 2539.871332
## iter 10 value 1324.127145
## iter 20 value 1072.120855
## iter 30 value 1050.418264
## iter 40 value 1049.691710
## iter 50 value 1047.865388
## iter 60 value 830.209056
## iter 70 value 660.299531
## iter 80 value 613.390646
## iter 90 value 599.172606
## iter 100 value 502.855616
## final value 502.855616
## stopped after 100 iterations
## # weights: 151
## initial value 1455.427716
## iter 10 value 1367.787514
## iter 20 value 1302.021602
## iter 30 value 1047.299473
## iter 40 value 751.555281
## iter 50 value 738.666842
## iter 60 value 632.837461
## iter 70 value 603.266246
## iter 80 value 564.864108
## iter 90 value 437.425772
## iter 100 value 397.424099
## final value 397.424099
## stopped after 100 iterations
## # weights: 46
## initial value 1716.615804
## iter 10 value 1330.691859
## iter 20 value 1324.978188
## iter 30 value 1324.901137
## iter 40 value 1291.558979
## iter 50 value 1097.602603
## iter 60 value 1039.078607
## iter 70 value 1020.443959
## iter 80 value 890.057397
## iter 90 value 706.729294
## iter 100 value 667.375865
## final value 667.375865
## stopped after 100 iterations
## # weights: 67
## initial value 2442.365932
## iter 10 value 1325.411565
## iter 20 value 1324.224202
## iter 30 value 1323.362062
## iter 40 value 1268.807116
## iter 50 value 1141.892172
## iter 60 value 1122.903912
## iter 70 value 920.875954
## iter 80 value 797.573150
## iter 90 value 754.961329
## iter 100 value 746.981560
## final value 746.981560
## stopped after 100 iterations
## # weights: 109
## initial value 1423.159828
## iter 10 value 1325.941609
## iter 20 value 1324.295618
## iter 30 value 1324.276786
## iter 40 value 1258.632223
## iter 50 value 1053.932840
## iter 60 value 1035.469108
## iter 70 value 889.854150
## iter 80 value 758.722848
## iter 90 value 714.341905
## iter 100 value 709.240407
## final value 709.240407
## stopped after 100 iterations
## # weights: 151
## initial value 2291.876913
## iter 10 value 1336.092839
## iter 20 value 1318.651428
## iter 30 value 913.541920
## iter 40 value 819.812832
## iter 50 value 732.139665
## iter 60 value 669.403275
## iter 70 value 610.860651
## iter 80 value 592.242694
## iter 90 value 584.361383
## iter 100 value 556.316027
## final value 556.316027
## stopped after 100 iterations
## # weights: 46
## initial value 1486.735337
## iter 10 value 1324.138373
## iter 20 value 977.506962
## iter 30 value 806.411289
## iter 40 value 785.558889
## iter 50 value 742.788564
## iter 60 value 741.256528
## iter 70 value 737.036355
## iter 80 value 731.733705
## iter 90 value 721.124192
## iter 100 value 475.363327
## final value 475.363327
## stopped after 100 iterations
## # weights: 67
## initial value 2332.645132
## iter 10 value 1324.385061
## iter 10 value 1324.385054
## iter 10 value 1324.385049
## final value 1324.385049
## converged
## # weights: 109
## initial value 1544.065159
## iter 10 value 1323.453242
## iter 20 value 1283.588166
## iter 30 value 1027.312682
## iter 40 value 899.995788
## iter 50 value 750.101451
## iter 60 value 744.722849
## iter 70 value 738.861463
## iter 80 value 736.084340
## iter 90 value 732.342903
## iter 100 value 702.524040
## final value 702.524040
## stopped after 100 iterations
## # weights: 151
## initial value 1763.316660
## iter 10 value 1434.937009
## iter 20 value 1349.484937
## iter 30 value 1309.754645
## iter 40 value 1175.079959
## iter 50 value 983.618472
## iter 60 value 807.875823
## iter 70 value 782.519808
## iter 80 value 742.783321
## iter 90 value 726.077165
## iter 100 value 657.852375
## final value 657.852375
## stopped after 100 iterations
## # weights: 46
## initial value 1465.634924
## iter 10 value 1323.308717
## iter 20 value 1248.128529
## iter 30 value 1096.906526
## iter 40 value 807.051673
## iter 50 value 737.689771
## iter 60 value 720.334165
## iter 70 value 716.406460
## iter 80 value 715.464178
## iter 90 value 510.040783
## iter 100 value 457.832834
## final value 457.832834
## stopped after 100 iterations
## # weights: 67
## initial value 1848.445548
## iter 10 value 1323.107241
## iter 20 value 1323.065819
## iter 30 value 1316.070968
## iter 40 value 1144.148111
## iter 50 value 1012.035411
## iter 60 value 909.872654
## iter 70 value 870.205213
## iter 80 value 861.952095
## iter 90 value 858.405523
## iter 100 value 841.795464
## final value 841.795464
## stopped after 100 iterations
## # weights: 109
## initial value 1395.510521
## iter 10 value 1322.844196
## iter 20 value 1048.810882
## iter 30 value 836.567284
## iter 40 value 776.743347
## iter 50 value 716.165948
## iter 60 value 693.278488
## iter 70 value 690.506627
## iter 80 value 687.508114
## iter 90 value 681.330098
## iter 100 value 527.397181
## final value 527.397181
## stopped after 100 iterations
## # weights: 151
## initial value 2145.077483
## iter 10 value 1325.396461
## iter 20 value 894.530230
## iter 30 value 825.155437
## iter 40 value 794.984749
## iter 50 value 780.645177
## iter 60 value 733.565213
## iter 70 value 657.081025
## iter 80 value 557.695656
## iter 90 value 532.691878
## iter 100 value 525.028669
## final value 525.028669
## stopped after 100 iterations
## # weights: 46
## initial value 1432.443096
## iter 10 value 1325.569383
## iter 20 value 1323.259914
## iter 30 value 1323.228229
## final value 1323.227858
## converged
## # weights: 67
## initial value 1793.871031
## iter 10 value 1324.792193
## iter 20 value 1323.248712
## iter 30 value 1323.228096
## iter 40 value 1289.715011
## iter 50 value 809.114463
## iter 60 value 759.647414
## iter 70 value 677.037385
## iter 80 value 588.608761
## iter 90 value 566.850157
## iter 100 value 558.771402
## final value 558.771402
## stopped after 100 iterations
## # weights: 109
## initial value 1593.185912
## iter 10 value 1327.654472
## iter 20 value 1322.836191
## iter 30 value 1279.675752
## iter 40 value 1168.296394
## iter 50 value 899.917035
## iter 60 value 861.805154
## iter 70 value 854.707854
## iter 80 value 797.389458
## iter 90 value 755.546698
## iter 100 value 699.209953
## final value 699.209953
## stopped after 100 iterations
## # weights: 151
## initial value 2036.054019
## iter 10 value 1329.999693
## iter 20 value 1323.127998
## iter 30 value 1322.203527
## iter 40 value 1225.678088
## iter 50 value 1129.874598
## iter 60 value 903.769688
## iter 70 value 743.452940
## iter 80 value 727.836355
## iter 90 value 722.905487
## iter 100 value 722.488203
## final value 722.488203
## stopped after 100 iterations
## # weights: 46
## initial value 1523.222362
## iter 10 value 1325.616422
## iter 20 value 1323.961527
## iter 30 value 1323.953504
## iter 30 value 1323.953493
## iter 40 value 1321.011637
## iter 50 value 1050.422460
## iter 60 value 938.888617
## iter 70 value 895.364925
## iter 80 value 865.160509
## iter 90 value 854.758175
## iter 100 value 845.934241
## final value 845.934241
## stopped after 100 iterations
## # weights: 67
## initial value 1420.205341
## iter 10 value 1323.733268
## iter 20 value 1298.050015
## iter 30 value 1126.852570
## iter 40 value 1104.191650
## iter 50 value 970.078126
## iter 60 value 856.776954
## iter 70 value 843.566358
## iter 80 value 761.892310
## iter 90 value 742.409610
## iter 100 value 737.722054
## final value 737.722054
## stopped after 100 iterations
## # weights: 109
## initial value 1705.124314
## iter 10 value 1321.750276
## iter 20 value 1084.077899
## iter 30 value 1071.447291
## iter 40 value 917.348001
## iter 50 value 818.563153
## iter 60 value 769.080308
## iter 70 value 758.393448
## iter 80 value 669.744702
## iter 90 value 652.156780
## iter 100 value 648.002289
## final value 648.002289
## stopped after 100 iterations
## # weights: 151
## initial value 1721.853758
## iter 10 value 1324.645121
## iter 20 value 1306.095781
## iter 30 value 1046.830827
## iter 40 value 810.693662
## iter 50 value 758.484166
## iter 60 value 742.368307
## iter 70 value 731.806172
## iter 80 value 714.704209
## iter 90 value 678.249652
## iter 100 value 527.536429
## final value 527.536429
## stopped after 100 iterations
## # weights: 46
## initial value 2008.671029
## iter 10 value 1321.836831
## iter 20 value 1305.142764
## iter 30 value 1171.122465
## iter 40 value 947.345564
## iter 50 value 756.131794
## iter 60 value 666.349809
## iter 70 value 487.839062
## iter 80 value 386.186885
## iter 90 value 335.069523
## iter 100 value 321.044118
## final value 321.044118
## stopped after 100 iterations
## # weights: 67
## initial value 1629.784319
## iter 10 value 1321.779828
## iter 20 value 1321.753483
## iter 30 value 1319.373598
## iter 40 value 1062.606443
## iter 50 value 888.588423
## iter 60 value 856.592518
## iter 70 value 767.326271
## iter 80 value 635.946450
## iter 90 value 454.706226
## iter 100 value 399.605601
## final value 399.605601
## stopped after 100 iterations
## # weights: 109
## initial value 1824.408888
## iter 10 value 1321.712439
## iter 20 value 1287.730012
## iter 30 value 1200.483029
## iter 40 value 1038.787762
## iter 50 value 748.628181
## iter 60 value 659.036516
## iter 70 value 633.143021
## iter 80 value 591.535245
## iter 90 value 476.640428
## iter 100 value 401.676973
## final value 401.676973
## stopped after 100 iterations
## # weights: 151
## initial value 1601.231261
## iter 10 value 1324.448390
## iter 20 value 1321.778933
## iter 30 value 1319.790812
## iter 40 value 1305.831757
## iter 50 value 1100.140531
## iter 60 value 780.404159
## iter 70 value 733.343902
## iter 80 value 607.322290
## iter 90 value 577.075140
## iter 100 value 563.349778
## final value 563.349778
## stopped after 100 iterations
## # weights: 46
## initial value 1895.174456
## iter 10 value 1322.581995
## iter 20 value 1322.003630
## iter 30 value 1321.996811
## iter 40 value 1298.529830
## iter 50 value 1019.477130
## iter 60 value 775.189155
## iter 70 value 731.556210
## iter 80 value 730.230373
## iter 90 value 699.570250
## iter 100 value 584.964075
## final value 584.964075
## stopped after 100 iterations
## # weights: 67
## initial value 1431.025344
## iter 10 value 1322.803446
## iter 20 value 1322.006539
## iter 30 value 1321.996845
## iter 40 value 1149.378901
## iter 50 value 1065.921644
## iter 60 value 1047.717399
## iter 70 value 922.031266
## iter 80 value 884.147592
## iter 90 value 871.705797
## iter 100 value 855.583292
## final value 855.583292
## stopped after 100 iterations
## # weights: 109
## initial value 1471.566041
## iter 10 value 1322.486084
## iter 20 value 1322.002375
## iter 30 value 1321.975675
## iter 40 value 1265.436157
## iter 50 value 814.952054
## iter 60 value 751.017923
## iter 70 value 727.346347
## iter 80 value 723.658935
## iter 90 value 722.556523
## iter 100 value 721.752502
## final value 721.752502
## stopped after 100 iterations
## # weights: 151
## initial value 1823.370950
## iter 10 value 1324.203147
## iter 20 value 1322.025449
## iter 30 value 1321.996470
## iter 40 value 1317.910914
## iter 50 value 1169.147310
## iter 60 value 1155.921445
## iter 70 value 866.675774
## iter 80 value 810.899488
## iter 90 value 744.491774
## iter 100 value 733.098765
## final value 733.098765
## stopped after 100 iterations
## # weights: 46
## initial value 1418.739965
## iter 10 value 1320.956817
## iter 20 value 1161.086619
## iter 30 value 1110.581933
## iter 40 value 892.544634
## iter 50 value 793.322794
## iter 60 value 751.833277
## iter 70 value 732.991426
## iter 80 value 696.287349
## iter 90 value 539.979137
## iter 100 value 360.631065
## final value 360.631065
## stopped after 100 iterations
## # weights: 67
## initial value 2132.083238
## iter 10 value 1322.774458
## iter 20 value 1322.766315
## iter 30 value 1322.733987
## iter 40 value 1322.726335
## iter 50 value 1321.971625
## iter 60 value 1319.445176
## iter 70 value 1302.761576
## iter 80 value 1300.850449
## iter 90 value 840.726686
## iter 100 value 780.362437
## final value 780.362437
## stopped after 100 iterations
## # weights: 109
## initial value 2073.734671
## iter 10 value 1324.841108
## iter 20 value 1285.710497
## iter 30 value 1164.607269
## iter 40 value 1077.633609
## iter 50 value 811.478788
## iter 60 value 745.971402
## iter 70 value 733.032855
## iter 80 value 730.540437
## iter 90 value 699.369829
## iter 100 value 591.500339
## final value 591.500339
## stopped after 100 iterations
## # weights: 151
## initial value 1461.912456
## iter 10 value 1321.829204
## iter 20 value 1129.371432
## iter 30 value 949.494409
## iter 40 value 867.667042
## iter 50 value 795.952913
## iter 60 value 776.571969
## iter 70 value 732.235832
## iter 80 value 728.406597
## iter 90 value 726.759742
## iter 100 value 695.629163
## final value 695.629163
## stopped after 100 iterations
## # weights: 109
## initial value 2169.478651
## iter 10 value 1983.975350
## iter 20 value 1827.227207
## iter 30 value 1222.218733
## iter 40 value 1125.029779
## iter 50 value 1063.266824
## iter 60 value 1054.155666
## iter 70 value 1014.031922
## iter 80 value 792.870596
## iter 90 value 667.532689
## iter 100 value 580.024116
## final value 580.024116
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n4_NN1Fit0
## Neural Network
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1061, 1060, 1059
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.8463874 0.7592319
## 2 0.5 0.6150948 0.3559886
## 2 0.7 0.8125665 0.7072849
## 3 0.3 0.8094242 0.7004414
## 3 0.5 0.6994249 0.5216280
## 3 0.7 0.6123972 0.3521753
## 5 0.3 0.8616064 0.7940362
## 5 0.5 0.7295603 0.5690515
## 5 0.7 0.7829199 0.6606979
## 7 0.3 0.8202108 0.7220827
## 7 0.5 0.7321176 0.5706245
## 7 0.7 0.7547139 0.6125326
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 5 and decay = 0.3.
DryBean_TDA_PC_5.50.5_n4_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.8487713 0.7734099 Fold1
## 2 0.8945386 0.8425234 Fold3
## 3 0.8415094 0.7661752 Fold2
db_tda_pc_5.50.5_n4_nn1_fit_re<-DryBean_TDA_PC_5.50.5_n4_NN1Fit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n4_NN1Fit0)
## a 16-5-4 network with 109 weights
## options were - softmax modelling decay=0.3
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 -0.15 -3.08 0.92 -0.48 0.01 0.00 0.18 0.10 0.01
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.03 0.01 1.93 -0.36 0.18 0.12 0.02 -0.03 -1.81 -0.33
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.02 -0.01 0.01 0.00 0.00 0.01 0.02
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 -0.02 0.00 0.00 0.00 0.00 0.00 -0.02 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1
## 0.23 -2.03 -0.04 0.23 0.23 0.00
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2
## -0.92 5.86 0.09 -0.94 -1.09 0.00
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3
## 1.23 -2.59 0.02 1.24 -2.93 0.00
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4
## -0.54 -1.23 -0.07 -0.53 3.79 0.00
#vip(DryBean_TDA_PC_5.50.5_n4_NN1Fit0,25) + ggtitle("dryBean_TDA_PCA_5.50.5_n4_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_PC_5.50.5_n4_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n4_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n4_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n4_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 145 0 418 0 7 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 251 0 71 1063 571 608 790
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2806
## 95% CI : (0.2669, 0.2947)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.001947
##
## Kappa : 0.1687
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.8548 0.0000
## Specificity 1.00000 1.00000 0.9577 1.0000
## Pos Pred Value NaN 1.00000 0.7333 NaN
## Neg Pred Value 0.90294 1.00000 0.9798 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.1025 0.0000
## Detection Prevalence 0.00000 0.03824 0.1397 0.0000
## Balanced Accuracy 0.50000 1.00000 0.9062 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9879 0.000 0.0000
## Specificity 0.2053 1.000 1.0000
## Pos Pred Value 0.1702 NaN NaN
## Neg Pred Value 0.9904 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1400 0.000 0.0000
## Detection Prevalence 0.8221 0.000 0.0000
## Balanced Accuracy 0.5966 0.500 0.5000
db_tda_pc_5.50.5_n4_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 145 0 418 0 7 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 251 0 71 1063 571 608 790
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2806
## 95% CI : (0.2669, 0.2947)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.001947
##
## Kappa : 0.1687
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.8548 0.0000
## Specificity 1.00000 1.00000 0.9577 1.0000
## Pos Pred Value NaN 1.00000 0.7333 NaN
## Neg Pred Value 0.90294 1.00000 0.9798 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.1025 0.0000
## Detection Prevalence 0.00000 0.03824 0.1397 0.0000
## Balanced Accuracy 0.50000 1.00000 0.9062 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9879 0.000 0.0000
## Specificity 0.2053 1.000 1.0000
## Pos Pred Value 0.1702 NaN NaN
## Neg Pred Value 0.9904 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1400 0.000 0.0000
## Detection Prevalence 0.8221 0.000 0.0000
## Balanced Accuracy 0.5966 0.500 0.5000
db_tda_pc_5.50.5_n4_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.28063725 0.16868927 0.26688891 0.29470037 0.26053922
## AccuracyPValue McnemarPValue
## 0.00194719 NaN
db_tda_pc_5.50.5_n4_db_nn1_cf0_ov_acc<-db_tda_pc_5.50.5_n4_db_nn1_cf0$overall[1]
db_tda_pc_5.50.5_n4_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.8548057 0.9576720 0.7333333 0.9797721 0.7333333
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9878893 0.2053113 0.1702445 0.9903581 0.1702445
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.00000000
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.8548057 0.7894240 0.11985294 0.10245098
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9878893 0.2904374 0.14166667 0.13995098
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.00000000 0.5000000
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.13970588 0.9062388
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.82205882 0.5966003
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
db_tda_pc_5.50.5_n4_db_nn1_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n4_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold<-(db_nn1_fit_re - db_tda_pc_5.50.5_n4_nn1_fit_re)
diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold
## Accuracy
## 1 -0.1431667
## 2 -0.3613312
## 3 -0.2741715
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold
## $winLeft
## [1] 0.9909333
##
## $winRope
## [1] 0.009066667
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n4_3_fold
## $left
## [1] 0.9618324
##
## $rope
## [1] 0.004922686
##
## $right
## [1] 0.03324494
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold))
#bf_tda_pca_5.50.5_nn1.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n4_3_fold)
## t = -4.0939, df = 2, p-value = 0.05481
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.53234974 0.01323681
## sample estimates:
## mean of x
## -0.2595565
### Test set diff
diff_drybean_tda_pca_5.50.5_nn1.n4_test<-(db_nn1_cf_ov_acc - db_tda_pc_5.50.5_n4_db_nn1_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nn1.n4_test
## Accuracy
## 0.3291667
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n4_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n4_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n4_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n4_test$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n4_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.164
##
## $winRight
## [1] 0.836
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n4_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nn1.n4_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n4_test)) #bf_tda_pca_5.50.5_nn1.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n4_test))
##Node5
#Neural Network 1
DryBean_TDA_PC_5.50.5_n5_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 37
## initial value 330.975925
## iter 10 value 10.529011
## iter 20 value 10.441081
## iter 30 value 7.781758
## iter 40 value 7.178670
## iter 50 value 6.512430
## iter 60 value 5.770999
## iter 70 value 5.766826
## iter 80 value 5.765927
## iter 90 value 5.765810
## final value 5.765790
## converged
## # weights: 55
## initial value 155.463032
## iter 10 value 12.076439
## iter 20 value 9.362283
## iter 30 value 9.362218
## iter 40 value 9.251706
## iter 50 value 8.639629
## iter 60 value 6.620488
## iter 70 value 5.299559
## iter 80 value 5.079599
## iter 90 value 4.977372
## iter 100 value 4.934320
## final value 4.934320
## stopped after 100 iterations
## # weights: 91
## initial value 250.386966
## iter 10 value 8.268968
## iter 20 value 7.077836
## iter 30 value 6.067821
## iter 40 value 5.850982
## iter 50 value 5.324141
## iter 60 value 4.242711
## iter 70 value 3.995921
## iter 80 value 3.966817
## iter 90 value 3.964395
## iter 100 value 3.909306
## final value 3.909306
## stopped after 100 iterations
## # weights: 127
## initial value 384.495175
## iter 10 value 11.395886
## iter 20 value 8.083242
## iter 30 value 7.715808
## iter 40 value 7.118165
## iter 50 value 5.963560
## iter 60 value 5.324041
## iter 70 value 5.053133
## iter 80 value 4.704147
## iter 90 value 4.474520
## iter 100 value 4.023932
## final value 4.023932
## stopped after 100 iterations
## # weights: 37
## initial value 429.463839
## iter 10 value 17.050565
## iter 20 value 11.064412
## iter 30 value 10.898644
## iter 40 value 10.898075
## iter 50 value 10.895718
## iter 60 value 10.719052
## iter 70 value 8.859840
## iter 80 value 7.676699
## iter 90 value 7.660207
## iter 100 value 7.656439
## final value 7.656439
## stopped after 100 iterations
## # weights: 55
## initial value 232.077831
## iter 10 value 11.735423
## iter 20 value 11.129678
## iter 30 value 10.263033
## iter 40 value 9.177586
## iter 50 value 8.496739
## iter 60 value 8.131594
## iter 70 value 7.650441
## iter 80 value 7.436428
## iter 90 value 7.309205
## iter 100 value 6.975349
## final value 6.975349
## stopped after 100 iterations
## # weights: 91
## initial value 701.316011
## iter 10 value 17.664352
## iter 20 value 17.586434
## iter 30 value 11.850141
## iter 40 value 8.951678
## iter 50 value 8.899924
## iter 60 value 8.809852
## iter 70 value 8.376801
## iter 80 value 7.731936
## iter 90 value 7.023255
## iter 100 value 6.126855
## final value 6.126855
## stopped after 100 iterations
## # weights: 127
## initial value 334.098119
## iter 10 value 22.987765
## iter 20 value 21.925718
## iter 30 value 21.184238
## iter 40 value 9.959539
## iter 50 value 9.167946
## iter 60 value 8.838899
## iter 70 value 7.951030
## iter 80 value 7.547414
## iter 90 value 6.876657
## iter 100 value 6.601309
## final value 6.601309
## stopped after 100 iterations
## # weights: 37
## initial value 318.872962
## iter 10 value 14.543180
## iter 20 value 14.450271
## iter 30 value 13.162107
## iter 40 value 12.378518
## iter 50 value 12.338011
## iter 60 value 12.329102
## iter 70 value 12.324819
## iter 80 value 12.294576
## iter 90 value 12.218869
## iter 100 value 11.258269
## final value 11.258269
## stopped after 100 iterations
## # weights: 55
## initial value 276.282907
## iter 10 value 13.097704
## iter 20 value 11.081413
## iter 30 value 10.859967
## iter 40 value 10.444283
## iter 50 value 7.773453
## iter 60 value 7.610163
## iter 70 value 7.600651
## iter 80 value 7.596402
## iter 90 value 7.595145
## iter 100 value 7.594995
## final value 7.594995
## stopped after 100 iterations
## # weights: 91
## initial value 182.469028
## iter 10 value 14.606902
## iter 20 value 14.508895
## iter 30 value 14.419357
## iter 40 value 12.287494
## iter 50 value 10.160711
## iter 60 value 9.320988
## iter 70 value 8.581432
## iter 80 value 8.311145
## iter 90 value 7.137264
## iter 100 value 6.720092
## final value 6.720092
## stopped after 100 iterations
## # weights: 127
## initial value 279.111549
## iter 10 value 11.774832
## iter 20 value 10.998805
## iter 30 value 9.910084
## iter 40 value 9.285195
## iter 50 value 9.174113
## iter 60 value 8.610866
## iter 70 value 7.773089
## iter 80 value 7.432637
## iter 90 value 7.137539
## iter 100 value 6.837281
## final value 6.837281
## stopped after 100 iterations
## # weights: 37
## initial value 301.641103
## iter 10 value 9.374783
## iter 20 value 9.369438
## iter 30 value 9.353473
## iter 40 value 9.244437
## iter 50 value 9.102951
## iter 60 value 7.273730
## iter 70 value 7.238786
## iter 80 value 7.229217
## iter 90 value 7.228270
## iter 100 value 7.228027
## final value 7.228027
## stopped after 100 iterations
## # weights: 55
## initial value 239.589545
## iter 10 value 13.604353
## iter 20 value 9.227744
## iter 30 value 8.750709
## iter 40 value 8.612530
## iter 50 value 7.926781
## iter 60 value 6.742459
## iter 70 value 5.532885
## iter 80 value 5.294815
## iter 90 value 4.696065
## iter 100 value 4.675240
## final value 4.675240
## stopped after 100 iterations
## # weights: 91
## initial value 62.791463
## iter 10 value 13.213617
## iter 20 value 8.298140
## iter 30 value 7.396596
## iter 40 value 7.122372
## iter 50 value 6.084309
## iter 60 value 5.514858
## iter 70 value 5.229513
## iter 80 value 4.956976
## iter 90 value 4.736555
## iter 100 value 4.338883
## final value 4.338883
## stopped after 100 iterations
## # weights: 127
## initial value 216.188245
## iter 10 value 7.892441
## iter 20 value 7.891364
## iter 30 value 7.371293
## iter 40 value 7.205767
## iter 50 value 6.407298
## iter 60 value 5.280510
## iter 70 value 4.918686
## iter 80 value 4.567606
## iter 90 value 4.338019
## iter 100 value 4.025482
## final value 4.025482
## stopped after 100 iterations
## # weights: 37
## initial value 148.912336
## iter 10 value 13.545524
## iter 20 value 12.141697
## iter 30 value 9.254637
## iter 40 value 9.110623
## iter 50 value 8.385853
## iter 60 value 7.683702
## iter 70 value 7.667398
## iter 80 value 7.663390
## iter 90 value 7.662521
## iter 100 value 7.662050
## final value 7.662050
## stopped after 100 iterations
## # weights: 55
## initial value 295.527735
## iter 10 value 12.616680
## iter 20 value 11.160131
## iter 30 value 8.494414
## iter 40 value 8.377531
## iter 50 value 7.843031
## iter 60 value 7.365818
## iter 70 value 7.041623
## iter 80 value 6.288786
## iter 90 value 6.261710
## iter 100 value 6.256490
## final value 6.256490
## stopped after 100 iterations
## # weights: 91
## initial value 166.601931
## iter 10 value 21.855608
## iter 20 value 10.702841
## iter 30 value 8.966772
## iter 40 value 8.743832
## iter 50 value 8.244048
## iter 60 value 7.574714
## iter 70 value 7.238070
## iter 80 value 6.712253
## iter 90 value 5.711994
## iter 100 value 5.494025
## final value 5.494025
## stopped after 100 iterations
## # weights: 127
## initial value 110.166876
## iter 10 value 9.285605
## iter 20 value 8.974528
## iter 30 value 8.956882
## iter 40 value 8.956627
## iter 50 value 8.800857
## iter 60 value 8.264602
## iter 70 value 7.321806
## iter 80 value 6.593819
## iter 90 value 6.177986
## iter 100 value 6.053944
## final value 6.053944
## stopped after 100 iterations
## # weights: 37
## initial value 253.202703
## iter 10 value 16.011492
## iter 20 value 12.777227
## iter 30 value 11.289253
## iter 40 value 10.671793
## iter 50 value 10.305795
## iter 60 value 9.332956
## iter 70 value 9.282944
## iter 80 value 9.280375
## iter 90 value 9.280054
## iter 100 value 9.279904
## final value 9.279904
## stopped after 100 iterations
## # weights: 55
## initial value 193.411083
## iter 10 value 13.122406
## iter 20 value 12.353991
## iter 30 value 11.770252
## iter 40 value 9.380961
## iter 50 value 8.557710
## iter 60 value 7.647395
## iter 70 value 7.607068
## iter 80 value 7.603644
## iter 90 value 7.602615
## iter 100 value 7.602102
## final value 7.602102
## stopped after 100 iterations
## # weights: 91
## initial value 180.811750
## iter 10 value 10.757291
## iter 20 value 10.079706
## iter 30 value 9.615206
## iter 40 value 9.444724
## iter 50 value 9.047579
## iter 60 value 8.578090
## iter 70 value 8.090973
## iter 80 value 7.688514
## iter 90 value 7.204510
## iter 100 value 6.601469
## final value 6.601469
## stopped after 100 iterations
## # weights: 127
## initial value 370.586297
## iter 10 value 12.636623
## iter 20 value 12.320300
## iter 30 value 11.963664
## iter 40 value 11.827853
## iter 50 value 10.719888
## iter 60 value 9.279646
## iter 70 value 9.070265
## iter 80 value 8.808843
## iter 90 value 8.436711
## iter 100 value 7.173954
## final value 7.173954
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 37
## initial value 289.803717
## iter 10 value 5.119959
## iter 20 value 4.154443
## final value 4.154443
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 55
## initial value 276.223193
## iter 10 value 5.708488
## iter 20 value 3.477317
## iter 30 value 3.374424
## iter 40 value 3.359989
## iter 50 value 3.359788
## final value 3.359788
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 91
## initial value 174.346234
## iter 10 value 7.445037
## iter 20 value 3.827063
## iter 30 value 2.481636
## iter 40 value 2.481044
## iter 50 value 2.481011
## final value 2.481010
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 127
## initial value 154.889377
## iter 10 value 6.642537
## iter 20 value 3.579117
## iter 30 value 2.213705
## iter 40 value 2.212138
## iter 50 value 2.209716
## iter 60 value 2.073360
## iter 70 value 1.996633
## iter 80 value 1.996285
## iter 90 value 1.995785
## iter 100 value 1.995481
## final value 1.995481
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 37
## initial value 323.520685
## iter 10 value 7.049902
## iter 20 value 6.018559
## final value 6.018557
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 55
## initial value 168.140057
## iter 10 value 10.136123
## iter 20 value 6.317699
## iter 30 value 4.985054
## iter 40 value 4.892131
## iter 50 value 4.891047
## iter 50 value 4.891047
## iter 60 value 4.889698
## final value 4.889678
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 91
## initial value 87.905872
## iter 10 value 15.339173
## iter 20 value 6.115757
## iter 30 value 4.596207
## iter 40 value 4.169809
## iter 50 value 3.709675
## iter 60 value 3.633435
## iter 70 value 3.632513
## iter 80 value 3.632504
## iter 90 value 3.632492
## iter 100 value 3.632471
## final value 3.632471
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 127
## initial value 444.964425
## iter 10 value 14.832112
## iter 20 value 4.725693
## iter 30 value 4.268668
## iter 40 value 3.963392
## iter 50 value 3.717371
## iter 60 value 3.356497
## iter 70 value 3.323176
## iter 80 value 3.313583
## iter 90 value 3.241582
## iter 100 value 3.241309
## final value 3.241309
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 37
## initial value 169.208781
## iter 10 value 8.584812
## iter 20 value 7.646229
## final value 7.646229
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 55
## initial value 189.898610
## iter 10 value 16.198008
## iter 20 value 6.317823
## iter 30 value 6.233678
## iter 40 value 6.232667
## iter 50 value 6.232655
## iter 50 value 6.232655
## iter 50 value 6.232655
## final value 6.232655
## converged
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 91
## initial value 197.680995
## iter 10 value 7.684905
## iter 20 value 6.283613
## iter 30 value 5.868712
## iter 40 value 5.645912
## iter 50 value 5.318940
## iter 60 value 5.308764
## iter 70 value 5.308563
## iter 80 value 5.308388
## iter 90 value 5.308345
## iter 100 value 4.662843
## final value 4.662843
## stopped after 100 iterations
## Warning in nnet.formula(.outcome ~ ., data = dat, size = param$size, decay =
## param$decay, : group 'CALI' is empty
## # weights: 127
## initial value 385.158316
## iter 10 value 8.568458
## iter 20 value 7.666189
## iter 30 value 7.012669
## iter 40 value 4.785432
## iter 50 value 4.664175
## iter 60 value 4.662060
## iter 70 value 4.167561
## iter 80 value 4.154567
## iter 90 value 3.792074
## iter 100 value 3.770756
## final value 3.770756
## stopped after 100 iterations
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## # weights: 37
## initial value 429.970293
## iter 10 value 13.590228
## iter 20 value 13.518697
## iter 30 value 13.086444
## iter 40 value 10.684221
## iter 50 value 10.327693
## iter 60 value 10.314934
## iter 70 value 10.311224
## iter 80 value 10.309263
## iter 90 value 10.308737
## iter 100 value 10.308436
## final value 10.308436
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n5_NN1Fit0
## Neural Network
##
## 417 samples
## 16 predictor
## 2 classes: 'BOMBAY', 'CALI'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 278, 279, 277
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.997619 0
## 2 0.5 0.997619 0
## 2 0.7 0.997619 0
## 3 0.3 0.997619 0
## 3 0.5 0.997619 0
## 3 0.7 0.997619 0
## 5 0.3 0.997619 0
## 5 0.5 0.997619 0
## 5 0.7 0.997619 0
## 7 0.3 0.997619 0
## 7 0.5 0.997619 0
## 7 0.7 0.997619 0
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 2 and decay = 0.7.
DryBean_TDA_PC_5.50.5_n5_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.9928571 0 Fold3
## 2 1.0000000 NA Fold2
## 3 1.0000000 NA Fold1
db_tda_pc_5.50.5_n5_nn1_fit_re<-DryBean_TDA_PC_5.50.5_n5_NN1Fit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n5_NN1Fit0)
## a 16-2-1 network with 37 weights
## options were - entropy fitting decay=0.7
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 -0.05 -0.02 -0.01 0.00 0.00 0.00 -0.01 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.00 -0.04 -0.02 0.00 0.00 0.00 0.00 -0.01 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o h1->o h2->o
## -1.31 -1.87 -1.87
#vip(DryBean_TDA_PC_5.50.5_n5_NN1Fit0,25) + ggtitle("dryBean_TDA_PCA_5.50.5_n5_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_PC_5.50.5_n5_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n5_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n5_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n5_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.03823529 0.00000000 0.03256139 0.04458199 0.26053922
## AccuracyPValue McnemarPValue
## 1.00000000 NaN
db_tda_pc_5.50.5_n5_db_nn1_cf0_ov_acc<-db_tda_pc_5.50.5_n5_db_nn1_cf0$overall[1]
db_tda_pc_5.50.5_n5_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0 1 NaN 0.9029412
## Class: BOMBAY 1 0 0.03823529 NaN
## Class: CALI 0 1 NaN 0.8801471
## Class: DERMASON 0 1 NaN 0.7394608
## Class: HOROZ 0 1 NaN 0.8583333
## Class: SEKER 0 1 NaN 0.8509804
## Class: SIRA 0 1 NaN 0.8063725
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA NA 0 NA 0.09705882 0.00000000
## Class: BOMBAY 0.03823529 1 0.07365439 0.03823529 0.03823529
## Class: CALI NA 0 NA 0.11985294 0.00000000
## Class: DERMASON NA 0 NA 0.26053922 0.00000000
## Class: HOROZ NA 0 NA 0.14166667 0.00000000
## Class: SEKER NA 0 NA 0.14901961 0.00000000
## Class: SIRA NA 0 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0 0.5
## Class: BOMBAY 1 0.5
## Class: CALI 0 0.5
## Class: DERMASON 0 0.5
## Class: HOROZ 0 0.5
## Class: SEKER 0 0.5
## Class: SIRA 0 0.5
db_tda_pc_5.50.5_n5_db_nn1_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n5_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold<-(db_nn1_fit_re - db_tda_pc_5.50.5_n5_nn1_fit_re)
diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold
## Accuracy
## 1 -0.2872526
## 2 -0.4667926
## 3 -0.4326621
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold
## $winLeft
## [1] 0.9926
##
## $winRope
## [1] 0.0074
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n5_3_fold
## $left
## [1] 0.9869414
##
## $rope
## [1] 0.001211897
##
## $right
## [1] 0.01184669
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold))
#bf_tda_pca_5.50.5_nn1.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nn1_n5_3_fold)
## t = -7.186, df = 2, p-value = 0.01882
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.6324178 -0.1587203
## sample estimates:
## mean of x
## -0.3955691
### Test set diff
diff_drybean_tda_pca_5.50.5_nn1.n5_test<-(db_nn1_cf_ov_acc - db_tda_pc_5.50.5_n5_db_nn1_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nn1.n5_test
## Accuracy
## 0.5715686
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n5_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nn1.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nn1.n5_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nn1.n5_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nn1.n5_test$probRight
bst_dbf_db_tda_pca_5.50.5_nn1.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nn1.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n5_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nn1.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1571
##
## $winRight
## [1] 0.8429
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nn1.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n5_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nn1.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nn1.n5_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nn1.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n5_test)) #bf_tda_pca_5.50.5_nn1.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nn1.n5_test))
##With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node1
#Neural Network 1
DryBean_TDA_KDE_5.50.5_n1_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n1.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 55
## initial value 11209.806029
## iter 10 value 10567.853418
## final value 10567.846799
## converged
## # weights: 79
## initial value 11131.014641
## iter 10 value 10568.045002
## iter 20 value 8920.687141
## iter 30 value 8573.018814
## iter 40 value 8270.148421
## iter 50 value 8207.008452
## iter 60 value 8153.264403
## iter 70 value 8141.517883
## iter 80 value 8122.431755
## iter 90 value 8087.332916
## iter 90 value 8087.332851
## final value 8087.332851
## converged
## # weights: 127
## initial value 12640.791580
## iter 10 value 10567.786135
## final value 10567.783340
## converged
## # weights: 175
## initial value 14611.007046
## iter 10 value 10595.637989
## iter 20 value 10567.481620
## iter 30 value 10083.358534
## iter 40 value 9823.370128
## iter 50 value 9702.557302
## iter 60 value 9667.040259
## iter 70 value 9509.163732
## iter 80 value 9485.075703
## iter 90 value 9334.225638
## iter 100 value 9092.395212
## final value 9092.395212
## stopped after 100 iterations
## # weights: 55
## initial value 12172.206983
## iter 10 value 10650.573540
## iter 20 value 10568.293324
## iter 30 value 10567.974924
## final value 10567.974689
## converged
## # weights: 79
## initial value 11980.627796
## iter 10 value 10574.427271
## iter 20 value 10567.948846
## iter 30 value 10567.868920
## final value 10567.868003
## converged
## # weights: 127
## initial value 13948.568881
## iter 10 value 10819.707676
## iter 20 value 10569.285767
## iter 30 value 10567.986521
## iter 40 value 10567.970115
## iter 50 value 10567.875306
## final value 10567.868771
## converged
## # weights: 175
## initial value 13260.639353
## iter 10 value 10580.472350
## iter 20 value 10507.172682
## iter 30 value 8913.632053
## iter 40 value 8353.889697
## iter 50 value 7148.135508
## iter 60 value 6868.547859
## iter 70 value 6454.403530
## iter 80 value 6044.123453
## iter 90 value 6008.893709
## iter 100 value 5630.857530
## final value 5630.857530
## stopped after 100 iterations
## # weights: 55
## initial value 11654.000970
## iter 10 value 10661.257865
## iter 20 value 10568.844049
## iter 30 value 10568.094218
## iter 40 value 10557.148108
## iter 50 value 10547.694920
## iter 60 value 10543.303613
## iter 70 value 8710.024238
## iter 80 value 8502.756876
## iter 90 value 8211.167963
## iter 100 value 8083.407656
## final value 8083.407656
## stopped after 100 iterations
## # weights: 79
## initial value 11922.894148
## iter 10 value 10571.528636
## iter 20 value 10568.029735
## iter 30 value 10567.957676
## iter 40 value 10567.881791
## final value 10567.878654
## converged
## # weights: 127
## initial value 11870.828835
## iter 10 value 10568.757049
## final value 10567.952799
## converged
## # weights: 175
## initial value 15388.204855
## iter 10 value 10651.811677
## iter 20 value 10566.563677
## iter 30 value 8873.736453
## iter 40 value 8787.364631
## iter 50 value 8781.197554
## iter 60 value 8680.217715
## iter 70 value 8473.739483
## iter 80 value 7694.753808
## iter 90 value 6776.116031
## iter 100 value 4677.700397
## final value 4677.700397
## stopped after 100 iterations
## # weights: 55
## initial value 12075.838161
## iter 10 value 10565.444067
## final value 10565.375333
## converged
## # weights: 79
## initial value 11862.031093
## iter 10 value 10565.393822
## final value 10565.375369
## converged
## # weights: 127
## initial value 10894.859108
## iter 10 value 10565.391241
## iter 20 value 10554.166758
## iter 30 value 10503.399737
## iter 40 value 10397.334455
## iter 50 value 9813.014348
## iter 60 value 7427.971813
## iter 70 value 6385.771106
## iter 80 value 5929.112047
## iter 90 value 5546.698981
## iter 100 value 5445.329815
## final value 5445.329815
## stopped after 100 iterations
## # weights: 175
## initial value 12395.444345
## iter 10 value 10565.311257
## iter 20 value 9962.847495
## iter 30 value 9275.780373
## iter 40 value 9117.746996
## iter 50 value 8618.941392
## iter 60 value 8533.556531
## iter 70 value 8472.086923
## iter 80 value 7432.620596
## iter 90 value 7098.533064
## iter 100 value 7022.632561
## final value 7022.632561
## stopped after 100 iterations
## # weights: 55
## initial value 11395.196601
## iter 10 value 10585.744174
## iter 20 value 10565.754495
## iter 30 value 10564.163485
## iter 40 value 9561.925669
## iter 50 value 9407.709209
## iter 60 value 9396.132481
## iter 70 value 9387.124978
## iter 80 value 9365.894433
## iter 90 value 9333.421092
## iter 100 value 9329.387456
## final value 9329.387456
## stopped after 100 iterations
## # weights: 79
## initial value 11917.780266
## iter 10 value 10578.937778
## iter 20 value 10565.575169
## iter 30 value 10564.744871
## iter 40 value 10164.027798
## iter 50 value 9931.585655
## iter 60 value 9504.967033
## iter 70 value 7456.413606
## iter 80 value 7346.350085
## iter 90 value 7185.825713
## iter 100 value 6837.609049
## final value 6837.609049
## stopped after 100 iterations
## # weights: 127
## initial value 10947.653512
## iter 10 value 10565.599759
## iter 20 value 10565.346444
## final value 10565.343643
## converged
## # weights: 175
## initial value 12562.792275
## iter 10 value 10581.891694
## iter 20 value 10565.597974
## iter 30 value 10565.398640
## iter 40 value 9837.330022
## iter 50 value 8604.602772
## iter 60 value 8324.085764
## iter 70 value 8265.975235
## iter 80 value 8215.154818
## iter 90 value 8031.767727
## iter 100 value 6559.429334
## final value 6559.429334
## stopped after 100 iterations
## # weights: 55
## initial value 12091.138715
## iter 10 value 10565.489621
## final value 10565.481389
## converged
## # weights: 79
## initial value 10948.498772
## iter 10 value 10565.553257
## iter 20 value 10561.722071
## iter 30 value 9446.289462
## iter 40 value 8653.421577
## iter 50 value 7520.786166
## iter 60 value 6191.426205
## iter 70 value 5461.086846
## iter 80 value 5105.772953
## iter 90 value 4699.609358
## iter 100 value 4104.991564
## final value 4104.991564
## stopped after 100 iterations
## # weights: 127
## initial value 11297.493816
## iter 10 value 10565.690499
## final value 10565.481422
## converged
## # weights: 175
## initial value 15744.526706
## iter 10 value 10887.415329
## iter 20 value 10622.443925
## iter 30 value 10508.684409
## iter 40 value 10356.234609
## iter 50 value 10156.354859
## iter 60 value 10022.122421
## iter 70 value 9875.421121
## iter 80 value 9763.204353
## iter 90 value 9699.411015
## iter 100 value 9599.646582
## final value 9599.646582
## stopped after 100 iterations
## # weights: 55
## initial value 11085.883751
## iter 10 value 10568.434391
## final value 10568.425176
## converged
## # weights: 79
## initial value 11402.514579
## iter 10 value 10568.675371
## iter 20 value 10568.384696
## final value 10568.361533
## converged
## # weights: 127
## initial value 12776.603845
## iter 10 value 10568.388557
## final value 10568.361711
## converged
## # weights: 175
## initial value 13575.991474
## iter 10 value 10629.264602
## iter 20 value 10568.396183
## iter 30 value 10568.345257
## iter 30 value 10568.345226
## final value 10568.344775
## converged
## # weights: 55
## initial value 11300.652842
## iter 10 value 10569.444907
## iter 20 value 10568.643844
## iter 30 value 10568.449422
## final value 10568.446405
## converged
## # weights: 79
## initial value 12327.081313
## iter 10 value 10570.811152
## iter 20 value 10568.421578
## iter 30 value 10568.393462
## iter 30 value 10568.393459
## iter 30 value 10568.393459
## final value 10568.393459
## converged
## # weights: 127
## initial value 14211.713692
## iter 10 value 10581.226212
## iter 20 value 10568.597422
## iter 30 value 10568.447531
## iter 30 value 10568.447459
## final value 10568.446363
## converged
## # weights: 175
## initial value 14851.340659
## iter 10 value 10577.132131
## iter 20 value 9596.464505
## iter 30 value 8282.657737
## iter 40 value 8020.089683
## iter 50 value 7846.403268
## iter 60 value 7801.754065
## iter 70 value 7362.769690
## iter 80 value 6853.428218
## iter 90 value 6554.363085
## iter 100 value 6159.018784
## final value 6159.018784
## stopped after 100 iterations
## # weights: 55
## initial value 12124.812782
## iter 10 value 10569.746909
## final value 10568.679044
## converged
## # weights: 79
## initial value 11363.058633
## iter 10 value 10569.173877
## iter 20 value 10038.497779
## iter 30 value 9081.024130
## iter 40 value 8400.182547
## iter 50 value 8350.410005
## iter 60 value 7783.839803
## iter 70 value 7407.491913
## iter 80 value 6887.480314
## iter 90 value 6530.553255
## iter 100 value 6381.048657
## final value 6381.048657
## stopped after 100 iterations
## # weights: 127
## initial value 12936.563546
## iter 10 value 10568.462684
## final value 10568.456574
## converged
## # weights: 175
## initial value 11122.074978
## iter 10 value 10568.610674
## iter 20 value 10568.458683
## iter 30 value 10267.717962
## iter 40 value 9849.017755
## iter 50 value 9422.939730
## iter 60 value 8602.200958
## iter 70 value 8239.148831
## iter 80 value 8226.140786
## iter 90 value 8185.358026
## iter 100 value 6435.668670
## final value 6435.668670
## stopped after 100 iterations
## # weights: 175
## initial value 21598.355112
## iter 10 value 16175.758406
## iter 20 value 15852.429090
## iter 30 value 15851.588795
## iter 40 value 15851.549105
## iter 40 value 15851.548954
## iter 50 value 15850.833864
## iter 60 value 15850.698456
## final value 15850.697746
## converged
DryBean_TDA_KDE_5.50.5_n1_NN1Fit0
## Neural Network
##
## 8473 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5649, 5648, 5649
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.2203470 0.00000000
## 2 0.5 0.2407600 0.02995836
## 2 0.7 0.2685056 0.06716065
## 3 0.3 0.2767682 0.07490392
## 3 0.5 0.3277216 0.14633841
## 3 0.7 0.5010936 0.37107435
## 5 0.3 0.3685476 0.19664149
## 5 0.5 0.2203470 0.00000000
## 5 0.7 0.2203470 0.00000000
## 7 0.3 0.3385889 0.16995984
## 7 0.5 0.5721713 0.46807421
## 7 0.7 0.5213290 0.41059802
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 7 and decay = 0.5.
DryBean_TDA_KDE_5.50.5_n1_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.5646018 0.4551806 Fold2
## 2 0.5984419 0.5078989 Fold1
## 3 0.5534703 0.4411431 Fold3
nb_tda_kde_5.50.5_n1_nn1_fit_re<-DryBean_TDA_KDE_5.50.5_n1_NN1Fit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n1_NN1Fit0)
## a 16-7-7 network with 175 weights
## options were - softmax modelling decay=0.5
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 -0.01 0.00 0.00 0.00 0.00 0.00 -0.01 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h6 i1->h6 i2->h6 i3->h6 i4->h6 i5->h6 i6->h6 i7->h6 i8->h6 i9->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h6 i11->h6 i12->h6 i13->h6 i14->h6 i15->h6 i16->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h7 i1->h7 i2->h7 i3->h7 i4->h7 i5->h7 i6->h7 i7->h7 i8->h7 i9->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h7 i11->h7 i12->h7 i13->h7 i14->h7 i15->h7 i16->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1 h6->o1 h7->o1
## 0.04 0.00 0.04 0.00 0.00 0.00 0.04 0.04
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2 h6->o2 h7->o2
## -0.19 0.00 -0.19 0.00 0.00 0.00 -0.19 -0.19
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3 h6->o3 h7->o3
## 0.09 0.00 0.10 0.00 0.00 0.00 0.10 0.10
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4 h6->o4 h7->o4
## 0.07 0.00 0.07 0.00 0.00 0.00 0.07 0.07
## b->o5 h1->o5 h2->o5 h3->o5 h4->o5 h5->o5 h6->o5 h7->o5
## 0.13 0.00 0.13 0.00 0.00 0.00 0.13 0.13
## b->o6 h1->o6 h2->o6 h3->o6 h4->o6 h5->o6 h6->o6 h7->o6
## -0.07 0.00 -0.07 0.00 0.00 0.00 -0.07 -0.07
## b->o7 h1->o7 h2->o7 h3->o7 h4->o7 h5->o7 h6->o7 h7->o7
## -0.08 0.00 -0.08 0.00 0.00 0.00 -0.08 -0.08
#vip(DryBean_TDA_KDE_5.50.5_n1_NN1Fit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n1_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n1_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n1_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n1_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
nb_tda_kde_5.50.5_n1_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 396 156 489 1063 578 608 790
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.1417
## 95% CI : (0.1311, 0.1527)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.0000
## Specificity 1.00000 1.00000 1.0000 1.0000
## Pos Pred Value NaN NaN NaN NaN
## Neg Pred Value 0.90294 0.96176 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.0000
## Detection Prevalence 0.00000 0.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 1.0000 0.000 0.0000
## Specificity 0.0000 1.000 1.0000
## Pos Pred Value 0.1417 NaN NaN
## Neg Pred Value NaN 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1417 0.000 0.0000
## Detection Prevalence 1.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
nb_tda_kde_5.50.5_n1_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 396 156 489 1063 578 608 790
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.1417
## 95% CI : (0.1311, 0.1527)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.0000
## Specificity 1.00000 1.00000 1.0000 1.0000
## Pos Pred Value NaN NaN NaN NaN
## Neg Pred Value 0.90294 0.96176 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.0000
## Detection Prevalence 0.00000 0.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 1.0000 0.000 0.0000
## Specificity 0.0000 1.000 1.0000
## Pos Pred Value 0.1417 NaN NaN
## Neg Pred Value NaN 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1417 0.000 0.0000
## Detection Prevalence 1.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
nb_tda_kde_5.50.5_n1_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.1416667 0.0000000 0.1311036 0.1527456 0.2605392
## AccuracyPValue McnemarPValue
## 1.0000000 NaN
nb_tda_kde_5.50.5_n1_db_nn1_cf0_ov_acc<-nb_tda_kde_5.50.5_n1_db_nn1_cf0$overall[1]
nb_tda_kde_5.50.5_n1_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0 1 NaN 0.9029412 NA
## Class: BOMBAY 0 1 NaN 0.9617647 NA
## Class: CALI 0 1 NaN 0.8801471 NA
## Class: DERMASON 0 1 NaN 0.7394608 NA
## Class: HOROZ 1 0 0.1416667 NaN 0.1416667
## Class: SEKER 0 1 NaN 0.8509804 NA
## Class: SIRA 0 1 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate Detection Prevalence
## Class: BARBUNYA 0 NA 0.09705882 0.0000000 0
## Class: BOMBAY 0 NA 0.03823529 0.0000000 0
## Class: CALI 0 NA 0.11985294 0.0000000 0
## Class: DERMASON 0 NA 0.26053922 0.0000000 0
## Class: HOROZ 1 0.2481752 0.14166667 0.1416667 1
## Class: SEKER 0 NA 0.14901961 0.0000000 0
## Class: SIRA 0 NA 0.19362745 0.0000000 0
## Balanced Accuracy
## Class: BARBUNYA 0.5
## Class: BOMBAY 0.5
## Class: CALI 0.5
## Class: DERMASON 0.5
## Class: HOROZ 0.5
## Class: SEKER 0.5
## Class: SIRA 0.5
nb_tda_kde_5.50.5_n1_db_nn1_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n1_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold<-(db_nn1_fit_re - nb_tda_kde_5.50.5_n1_nn1_fit_re)
diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold
## Accuracy
## 1 0.14100276
## 2 -0.06523450
## 3 0.01386769
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n1_3_fold
## $probLeft
## [1] 0.25
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n1_3_fold_odds.left<-bst_tda_kde_5.50.5_nn1.n1_3_fold$probLeft/bst_tda_kde_5.50.5_nn1.n1_3_fold$probRight
bst_tda_kde_5.50.5_nn1.n1_3_fold_odds.left
## [1] 0.5
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n1_3_fold
## $winLeft
## [1] 0.2772
##
## $winRope
## [1] 0.05986667
##
## $winRight
## [1] 0.6629333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n1_3_fold
## $left
## [1] 0.3117017
##
## $rope
## [1] 0.08899445
##
## $right
## [1] 0.5993038
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold))
#bf_tda_kde_5.50.5_nn1.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n1_3_fold)
## t = 0.49739, df = 2, p-value = 0.6682
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.2285877 0.2883450
## sample estimates:
## mean of x
## 0.02987865
### Test set diff
diff_drybean_tda_kde_5.50.5_nn1.n1_test<-(db_nn1_cf_ov_acc - nb_tda_kde_5.50.5_n1_db_nn1_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nn1.n1_test
## Accuracy
## 0.4681373
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n1_test),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n1_test_odds.left<-bst_tda_kde_5.50.5_nn1.n1_test$probLeft/bst_tda_kde_5.50.5_nn1.n1_test$probRight
bst_tda_kde_5.50.5_nn1.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n1_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1603333
##
## $winRight
## [1] 0.8396667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n1_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nn1.n1_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n1_test)) #bf_tda_pca_5.50.5_nn1.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n1_test))
##Node2
#Neural Network 1
DryBean_TDA_KDE_5.50.5_n2_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n2.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 52
## initial value 8556.797540
## iter 10 value 8279.624882
## iter 20 value 6598.824199
## iter 30 value 6305.531459
## iter 40 value 6242.857077
## iter 50 value 6119.721398
## iter 60 value 5996.561258
## iter 70 value 5931.139467
## iter 80 value 5712.401497
## iter 90 value 5511.497037
## iter 100 value 4867.343811
## final value 4867.343811
## stopped after 100 iterations
## # weights: 75
## initial value 10017.436483
## iter 10 value 8279.126532
## iter 20 value 8212.426674
## iter 30 value 8002.608796
## iter 40 value 6266.657248
## iter 50 value 5861.176678
## iter 60 value 5566.949541
## iter 70 value 5481.181744
## iter 80 value 5457.009635
## iter 90 value 5414.394210
## iter 100 value 5320.855786
## final value 5320.855786
## stopped after 100 iterations
## # weights: 121
## initial value 9910.648529
## iter 10 value 8279.071006
## final value 8279.061546
## converged
## # weights: 167
## initial value 10451.094004
## iter 10 value 8279.093590
## iter 20 value 8180.589562
## iter 30 value 7705.175800
## iter 40 value 7605.193300
## iter 50 value 6519.057468
## iter 60 value 6497.787981
## iter 70 value 5916.344956
## iter 80 value 5165.611323
## iter 90 value 4944.099149
## iter 100 value 4670.162999
## final value 4670.162999
## stopped after 100 iterations
## # weights: 52
## initial value 12460.881773
## iter 10 value 9501.487927
## iter 20 value 8312.505094
## iter 30 value 6390.088366
## iter 40 value 6254.451593
## iter 50 value 5848.999732
## iter 60 value 4894.979821
## iter 70 value 4558.021749
## iter 80 value 4498.368804
## iter 90 value 4436.739144
## iter 100 value 4340.937975
## final value 4340.937975
## stopped after 100 iterations
## # weights: 75
## initial value 9624.012249
## iter 10 value 8281.307179
## iter 20 value 8279.314584
## iter 30 value 8279.291008
## iter 30 value 8279.290940
## iter 30 value 8279.290872
## final value 8279.290872
## converged
## # weights: 121
## initial value 9643.699985
## iter 10 value 8282.247008
## iter 20 value 8279.324647
## iter 30 value 8279.291171
## final value 8279.290851
## converged
## # weights: 167
## initial value 9496.738010
## iter 10 value 8280.238729
## iter 20 value 8279.118937
## final value 8279.068432
## converged
## # weights: 52
## initial value 9417.552926
## iter 10 value 8319.948539
## iter 20 value 8280.536582
## iter 30 value 8207.375165
## iter 40 value 7670.404322
## iter 50 value 7546.202876
## iter 60 value 6854.682866
## iter 70 value 5603.876571
## iter 80 value 5387.299231
## iter 90 value 4962.953258
## iter 100 value 4690.757640
## final value 4690.757640
## stopped after 100 iterations
## # weights: 75
## initial value 8897.185190
## iter 10 value 8279.751511
## final value 8279.748874
## converged
## # weights: 121
## initial value 11203.197903
## iter 10 value 8300.898629
## iter 20 value 8279.209543
## iter 30 value 8270.526167
## iter 40 value 8149.600654
## iter 50 value 6664.470253
## iter 60 value 6323.344840
## iter 70 value 4694.337756
## iter 80 value 4216.502549
## iter 90 value 4156.612240
## iter 100 value 3933.298242
## final value 3933.298242
## stopped after 100 iterations
## # weights: 167
## initial value 11710.534120
## iter 10 value 8279.306394
## iter 20 value 7615.140736
## iter 30 value 7534.152586
## iter 40 value 7024.252946
## iter 50 value 5796.410046
## iter 60 value 5176.412565
## iter 70 value 4989.753249
## iter 80 value 4441.219589
## iter 90 value 3704.837932
## iter 100 value 3399.020079
## final value 3399.020079
## stopped after 100 iterations
## # weights: 52
## initial value 10554.368374
## iter 10 value 8358.403765
## iter 20 value 8278.496491
## iter 30 value 8173.861229
## iter 40 value 7237.392537
## iter 50 value 7002.893507
## iter 60 value 5978.507746
## iter 70 value 3344.440488
## iter 80 value 2782.944170
## iter 90 value 2417.881325
## iter 100 value 2276.767709
## final value 2276.767709
## stopped after 100 iterations
## # weights: 75
## initial value 9891.882718
## iter 10 value 8276.001422
## final value 8275.999850
## converged
## # weights: 121
## initial value 10360.497289
## iter 10 value 8275.939631
## final value 8275.937293
## converged
## # weights: 167
## initial value 10173.016202
## iter 10 value 8320.319780
## iter 20 value 8278.466633
## iter 30 value 8277.922237
## iter 40 value 6496.284046
## iter 50 value 6284.827479
## iter 60 value 6114.007203
## iter 70 value 5523.089400
## iter 80 value 5070.041380
## iter 90 value 4135.547558
## iter 100 value 3662.258021
## final value 3662.258021
## stopped after 100 iterations
## # weights: 52
## initial value 9056.782007
## iter 10 value 8278.354986
## iter 20 value 8276.192090
## iter 30 value 8276.166379
## iter 30 value 8276.166365
## iter 30 value 8276.166363
## final value 8276.166363
## converged
## # weights: 75
## initial value 11347.541623
## iter 10 value 8370.879806
## iter 20 value 8279.793427
## iter 30 value 8278.105301
## iter 40 value 8276.189735
## iter 50 value 8272.856101
## iter 60 value 7973.197066
## iter 70 value 6259.859845
## iter 80 value 5962.820574
## iter 90 value 5920.042962
## iter 100 value 5883.614356
## final value 5883.614356
## stopped after 100 iterations
## # weights: 121
## initial value 11345.935240
## iter 10 value 8294.272906
## iter 20 value 8276.283313
## iter 30 value 8276.063716
## iter 40 value 8275.993346
## final value 8275.957950
## converged
## # weights: 167
## initial value 10481.572948
## iter 10 value 8294.669530
## iter 20 value 8276.247954
## iter 30 value 8276.021999
## iter 40 value 8275.803364
## iter 50 value 7901.830896
## iter 60 value 6207.066121
## iter 70 value 6078.317528
## iter 80 value 6043.871686
## iter 90 value 5992.725517
## iter 100 value 5934.513441
## final value 5934.513441
## stopped after 100 iterations
## # weights: 52
## initial value 8781.201522
## iter 10 value 8276.657878
## final value 8276.628163
## converged
## # weights: 75
## initial value 9304.170538
## iter 10 value 8276.722856
## iter 20 value 8276.624965
## final value 8276.624142
## converged
## # weights: 121
## initial value 9934.907792
## iter 10 value 8276.211732
## iter 20 value 8276.187370
## final value 8276.187223
## converged
## # weights: 167
## initial value 10873.715541
## iter 10 value 8277.145816
## iter 20 value 7289.105388
## iter 30 value 6046.558372
## iter 40 value 5974.914053
## iter 50 value 5558.132997
## iter 60 value 5202.733557
## iter 70 value 4672.990352
## iter 80 value 4482.058892
## iter 90 value 4457.524904
## iter 100 value 4437.704155
## final value 4437.704155
## stopped after 100 iterations
## # weights: 52
## initial value 9442.925613
## iter 10 value 8276.500008
## iter 20 value 8276.291949
## iter 30 value 8067.508394
## iter 40 value 7087.199427
## iter 50 value 6423.534584
## iter 60 value 6049.238469
## iter 70 value 5928.664256
## iter 80 value 5873.971112
## iter 90 value 5730.404031
## iter 100 value 5519.770201
## final value 5519.770201
## stopped after 100 iterations
## # weights: 75
## initial value 10302.819421
## iter 10 value 8282.658707
## iter 20 value 8222.476982
## iter 30 value 6725.028033
## iter 40 value 5995.028875
## iter 50 value 5891.091492
## iter 60 value 5849.710524
## iter 70 value 5840.506783
## iter 80 value 5836.707018
## iter 90 value 5792.903999
## iter 100 value 5454.205802
## final value 5454.205802
## stopped after 100 iterations
## # weights: 121
## initial value 10713.913050
## iter 10 value 8275.963922
## iter 20 value 8275.937220
## iter 20 value 8275.937146
## iter 20 value 8275.937142
## final value 8275.937142
## converged
## # weights: 167
## initial value 12111.374322
## iter 10 value 8388.716582
## iter 20 value 8276.509362
## iter 30 value 6883.112287
## iter 40 value 6662.634893
## iter 50 value 6618.938407
## iter 60 value 6128.732653
## iter 70 value 4617.497993
## iter 80 value 4175.655601
## iter 90 value 4085.361442
## iter 100 value 3952.972145
## final value 3952.972145
## stopped after 100 iterations
## # weights: 52
## initial value 8833.220207
## iter 10 value 8304.931535
## iter 20 value 8276.832713
## iter 30 value 8276.380053
## iter 40 value 8276.287657
## final value 8276.213883
## converged
## # weights: 75
## initial value 9529.981318
## iter 10 value 8288.414527
## iter 20 value 8276.346441
## iter 30 value 8276.168452
## iter 40 value 8276.161502
## iter 50 value 8276.064058
## final value 8276.062140
## converged
## # weights: 121
## initial value 10728.391818
## iter 10 value 8285.004765
## iter 20 value 8276.275219
## iter 30 value 8276.166482
## iter 40 value 8252.707410
## iter 50 value 7920.646148
## iter 60 value 6872.086826
## iter 70 value 6815.211211
## iter 80 value 6711.126940
## iter 90 value 6588.789633
## iter 100 value 5529.651982
## final value 5529.651982
## stopped after 100 iterations
## # weights: 167
## initial value 11249.374020
## iter 10 value 8286.713581
## iter 20 value 8276.210709
## iter 30 value 7894.554621
## iter 40 value 5248.269991
## iter 50 value 4712.548515
## iter 60 value 4507.393629
## iter 70 value 4423.484066
## iter 80 value 4304.432429
## iter 90 value 4219.297665
## iter 100 value 4084.950417
## final value 4084.950417
## stopped after 100 iterations
## # weights: 52
## initial value 8969.503431
## iter 10 value 8281.273629
## iter 20 value 8277.475032
## iter 30 value 8276.345315
## final value 8276.333173
## converged
## # weights: 75
## initial value 9973.482086
## iter 10 value 8276.631032
## final value 8276.624058
## converged
## # weights: 121
## initial value 11565.747785
## iter 10 value 8283.866391
## iter 20 value 8276.113673
## iter 30 value 8276.051979
## iter 40 value 8275.877463
## iter 50 value 6383.416083
## iter 60 value 6218.862022
## iter 70 value 6115.945912
## iter 80 value 5612.295198
## iter 90 value 5009.156735
## iter 100 value 3730.734991
## final value 3730.734991
## stopped after 100 iterations
## # weights: 167
## initial value 8645.997056
## iter 10 value 8222.015671
## iter 20 value 7987.332860
## iter 30 value 7474.042521
## iter 40 value 7273.091666
## iter 50 value 7155.246662
## iter 60 value 6918.447118
## iter 70 value 6222.874484
## iter 80 value 5869.183448
## iter 90 value 5711.488647
## iter 100 value 3835.679055
## final value 3835.679055
## stopped after 100 iterations
## # weights: 167
## initial value 16686.819595
## iter 10 value 12447.576631
## iter 20 value 12416.015396
## iter 30 value 12415.716637
## iter 40 value 10445.026952
## iter 50 value 10064.406715
## iter 60 value 9928.591552
## iter 70 value 9771.328477
## iter 80 value 7931.544590
## iter 90 value 7275.661317
## iter 100 value 6970.445498
## final value 6970.445498
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n2_NN1Fit0
## Neural Network
##
## 7582 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5056, 5054, 5054
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.6706414 0.5569011
## 2 0.5 0.4217263 0.1907494
## 2 0.7 0.4202747 0.1891611
## 3 0.3 0.4543885 0.2352842
## 3 0.5 0.3742853 0.1210137
## 3 0.7 0.2896334 0.0000000
## 5 0.3 0.2896334 0.0000000
## 5 0.5 0.3463317 0.1064236
## 5 0.7 0.5798279 0.4266460
## 7 0.3 0.7098207 0.6172923
## 7 0.5 0.5172178 0.3381399
## 7 0.7 0.7461163 0.6661104
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 7 and decay = 0.7.
DryBean_TDA_KDE_5.50.5_n2_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.6843354 0.5708785 Fold2
## 2 0.7731591 0.7090646 Fold1
## 3 0.7808544 0.7183882 Fold3
nb_tda_kde_5.50.5_n2_nn1_fit_re<-DryBean_TDA_KDE_5.50.5_n2_NN1Fit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n2_NN1Fit0)
## a 16-7-6 network with 167 weights
## options were - softmax modelling decay=0.7
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.05 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## -0.03 -1.21 -8.88 -2.52 -3.62 -0.03 -0.01 1.42 -3.10 -0.02
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## -0.03 -0.03 -0.03 0.00 0.00 -0.03 -0.03
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.06 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h6 i1->h6 i2->h6 i3->h6 i4->h6 i5->h6 i6->h6 i7->h6 i8->h6 i9->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h6 i11->h6 i12->h6 i13->h6 i14->h6 i15->h6 i16->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h7 i1->h7 i2->h7 i3->h7 i4->h7 i5->h7 i6->h7 i7->h7 i8->h7 i9->h7
## 0.02 0.76 2.86 -2.41 4.87 -0.02 0.00 -0.81 2.08 0.00
## i10->h7 i11->h7 i12->h7 i13->h7 i14->h7 i15->h7 i16->h7
## 0.02 0.03 0.03 0.00 0.00 0.03 0.02
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1 h6->o1 h7->o1
## 0.25 0.36 0.79 0.25 0.14 0.00 0.00 -3.01
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2 h6->o2 h7->o2
## 0.71 0.18 0.85 0.71 0.11 0.00 0.00 -5.71
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3 h6->o3 h7->o3
## -0.30 0.38 -4.42 -0.30 -0.35 0.00 0.00 4.74
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4 h6->o4 h7->o4
## 0.71 0.30 1.01 0.71 -0.25 0.00 0.00 -2.34
## b->o5 h1->o5 h2->o5 h3->o5 h4->o5 h5->o5 h6->o5 h7->o5
## -0.84 -0.41 -1.18 -0.84 0.74 0.00 0.00 4.44
## b->o6 h1->o6 h2->o6 h3->o6 h4->o6 h5->o6 h6->o6 h7->o6
## -0.53 -0.80 2.95 -0.53 -0.39 0.00 0.00 1.88
#vip(DryBean_TDA_KDE_5.50.5_n2_NN1Fit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n2_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n2_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n2_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n2_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n2_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 381 156 484 1 420 4 44
## DERMASON 0 0 0 911 24 216 52
## HOROZ 0 0 0 0 1 0 0
## SEKER 0 0 0 1 0 2 0
## SIRA 15 0 5 150 133 386 694
##
## Overall Statistics
##
## Accuracy : 0.5127
## 95% CI : (0.4973, 0.5282)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4011
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.9898 0.8570
## Specificity 1.00000 1.00000 0.7199 0.9032
## Pos Pred Value NaN NaN 0.3248 0.7573
## Neg Pred Value 0.90294 0.96176 0.9981 0.9472
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.1186 0.2233
## Detection Prevalence 0.00000 0.00000 0.3652 0.2949
## Balanced Accuracy 0.50000 0.50000 0.8548 0.8801
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0017301 0.0032895 0.8785
## Specificity 1.0000000 0.9997120 0.7906
## Pos Pred Value 1.0000000 0.6666667 0.5018
## Neg Pred Value 0.8585438 0.8513613 0.9644
## Prevalence 0.1416667 0.1490196 0.1936
## Detection Rate 0.0002451 0.0004902 0.1701
## Detection Prevalence 0.0002451 0.0007353 0.3390
## Balanced Accuracy 0.5008651 0.5015007 0.8345
nb_tda_kde_5.50.5_n2_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 381 156 484 1 420 4 44
## DERMASON 0 0 0 911 24 216 52
## HOROZ 0 0 0 0 1 0 0
## SEKER 0 0 0 1 0 2 0
## SIRA 15 0 5 150 133 386 694
##
## Overall Statistics
##
## Accuracy : 0.5127
## 95% CI : (0.4973, 0.5282)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4011
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.9898 0.8570
## Specificity 1.00000 1.00000 0.7199 0.9032
## Pos Pred Value NaN NaN 0.3248 0.7573
## Neg Pred Value 0.90294 0.96176 0.9981 0.9472
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.1186 0.2233
## Detection Prevalence 0.00000 0.00000 0.3652 0.2949
## Balanced Accuracy 0.50000 0.50000 0.8548 0.8801
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0017301 0.0032895 0.8785
## Specificity 1.0000000 0.9997120 0.7906
## Pos Pred Value 1.0000000 0.6666667 0.5018
## Neg Pred Value 0.8585438 0.8513613 0.9644
## Prevalence 0.1416667 0.1490196 0.1936
## Detection Rate 0.0002451 0.0004902 0.1701
## Detection Prevalence 0.0002451 0.0007353 0.3390
## Balanced Accuracy 0.5008651 0.5015007 0.8345
nb_tda_kde_5.50.5_n2_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.127451e-01 4.011354e-01 4.972808e-01 5.281911e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 2.042121e-257 NaN
nb_tda_kde_5.50.5_n2_db_nn1_cf0_ov_acc<-nb_tda_kde_5.50.5_n2_db_nn1_cf0$overall[1]
nb_tda_kde_5.50.5_n2_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.000000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.000000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.989775051 0.7198552 0.3248322 0.9980695 0.3248322
## Class: DERMASON 0.857008467 0.9032151 0.7572735 0.9471672 0.7572735
## Class: HOROZ 0.001730104 1.0000000 1.0000000 0.8585438 1.0000000
## Class: SEKER 0.003289474 0.9997120 0.6666667 0.8513613 0.6666667
## Class: SIRA 0.878481013 0.7905775 0.5018077 0.9644049 0.5018077
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.000000000 NA 0.09705882 0.0000000000
## Class: BOMBAY 0.000000000 NA 0.03823529 0.0000000000
## Class: CALI 0.989775051 0.489135927 0.11985294 0.1186274510
## Class: DERMASON 0.857008467 0.804060018 0.26053922 0.2232843137
## Class: HOROZ 0.001730104 0.003454231 0.14166667 0.0002450980
## Class: SEKER 0.003289474 0.006546645 0.14901961 0.0004901961
## Class: SIRA 0.878481013 0.638748274 0.19362745 0.1700980392
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000000 0.5000000
## Class: BOMBAY 0.0000000000 0.5000000
## Class: CALI 0.3651960784 0.8548151
## Class: DERMASON 0.2948529412 0.8801118
## Class: HOROZ 0.0002450980 0.5008651
## Class: SEKER 0.0007352941 0.5015007
## Class: SIRA 0.3389705882 0.8345293
nb_tda_kde_5.50.5_n2_db_nn1_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n2_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold<-(db_nn1_fit_re - nb_tda_kde_5.50.5_n2_nn1_fit_re)
diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold
## Accuracy
## 1 0.02126909
## 2 -0.23995172
## 3 -0.21351648
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n2_3_fold
## $probLeft
## [1] 0.5
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.25
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n2_3_fold_odds.left<-bst_tda_kde_5.50.5_nn1.n2_3_fold$probLeft/bst_tda_kde_5.50.5_nn1.n2_3_fold$probRight
bst_tda_kde_5.50.5_nn1.n2_3_fold_odds.left
## [1] 2
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n2_3_fold
## $winLeft
## [1] 0.8794
##
## $winRope
## [1] 0.01496667
##
## $winRight
## [1] 0.1056333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n2_3_fold
## $left
## [1] 0.8515766
##
## $rope
## [1] 0.02379056
##
## $right
## [1] 0.1246328
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold))
#bf_tda_kde_5.50.5_nn1.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n2_3_fold)
## t = -1.7353, df = 2, p-value = 0.2248
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.5012692 0.2131364
## sample estimates:
## mean of x
## -0.1440664
### Test set diff
diff_drybean_tda_kde_5.50.5_nn1.n2_test<-(db_nn1_cf_ov_acc - nb_tda_kde_5.50.5_n2_db_nn1_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nn1.n2_test
## Accuracy
## 0.09705882
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n2_test),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n2_test_odds.left<-bst_tda_kde_5.50.5_nn1.n2_test$probLeft/bst_tda_kde_5.50.5_nn1.n2_test$probRight
bst_tda_kde_5.50.5_nn1.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n2_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1643
##
## $winRight
## [1] 0.8357
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n2_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nn1.n2_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n2_test)) #bf_tda_pca_5.50.5_nn1.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n2_test))
##Node3
#Neural Network 1
DryBean_TDA_KDE_5.50.5_n3_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n3.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 52
## initial value 4969.978330
## iter 10 value 3209.964105
## iter 20 value 3177.749923
## iter 30 value 3120.615957
## iter 40 value 3103.737714
## iter 50 value 2602.039848
## iter 60 value 2282.226082
## iter 70 value 2071.494370
## iter 80 value 1972.288627
## iter 90 value 1658.636916
## iter 100 value 1338.700583
## final value 1338.700583
## stopped after 100 iterations
## # weights: 75
## initial value 4768.323946
## iter 10 value 3218.230308
## iter 20 value 3208.790199
## iter 30 value 3197.101123
## iter 40 value 2534.125828
## iter 50 value 2349.468334
## iter 60 value 2133.959554
## iter 70 value 2077.852058
## iter 80 value 2069.169026
## iter 90 value 1985.358537
## iter 100 value 1736.405320
## final value 1736.405320
## stopped after 100 iterations
## # weights: 121
## initial value 6994.457579
## iter 10 value 3226.320985
## iter 20 value 3218.684202
## iter 30 value 3217.030364
## iter 40 value 3216.370855
## iter 50 value 2477.359225
## iter 60 value 2341.398950
## iter 70 value 2278.545360
## iter 80 value 2150.475188
## iter 90 value 2094.190053
## iter 100 value 2032.248977
## final value 2032.248977
## stopped after 100 iterations
## # weights: 167
## initial value 6750.289785
## iter 10 value 3299.803199
## iter 20 value 3219.802074
## iter 30 value 3205.865098
## iter 40 value 3205.776043
## iter 50 value 3203.463996
## iter 60 value 2541.376338
## iter 70 value 2290.492178
## iter 80 value 2280.137994
## iter 90 value 1950.652845
## iter 100 value 1791.627453
## final value 1791.627453
## stopped after 100 iterations
## # weights: 52
## initial value 5526.861372
## iter 10 value 3324.074827
## iter 20 value 3258.434455
## iter 30 value 3224.745788
## iter 40 value 3210.762897
## iter 50 value 3208.858908
## iter 60 value 3208.823561
## final value 3208.822684
## converged
## # weights: 75
## initial value 5347.478881
## iter 10 value 3229.916593
## iter 20 value 3218.524211
## iter 30 value 3207.411904
## iter 40 value 2821.331417
## iter 50 value 2350.708567
## iter 60 value 1864.339228
## iter 70 value 1555.099637
## iter 80 value 1203.672709
## iter 90 value 1087.293173
## iter 100 value 1018.467097
## final value 1018.467097
## stopped after 100 iterations
## # weights: 121
## initial value 5947.264068
## iter 10 value 3303.343551
## iter 20 value 3263.877861
## iter 30 value 3222.245384
## iter 40 value 3206.747832
## iter 50 value 3206.634228
## iter 60 value 3206.598171
## iter 70 value 3206.594573
## iter 70 value 3206.594547
## iter 70 value 3206.594530
## final value 3206.594530
## converged
## # weights: 167
## initial value 5515.691524
## iter 10 value 3212.477546
## iter 20 value 3209.883947
## iter 30 value 3207.700952
## iter 40 value 3047.366315
## iter 50 value 2029.529104
## iter 60 value 2018.537335
## iter 70 value 1994.122916
## iter 80 value 1922.771602
## iter 90 value 1647.636170
## iter 100 value 1353.537010
## final value 1353.537010
## stopped after 100 iterations
## # weights: 52
## initial value 5007.327551
## iter 10 value 3222.650255
## iter 20 value 3211.514901
## iter 30 value 2532.008407
## iter 40 value 2216.554554
## iter 50 value 2084.466101
## iter 60 value 2040.677180
## iter 70 value 1935.597925
## iter 80 value 1761.614038
## iter 90 value 1759.642357
## final value 1759.640793
## converged
## # weights: 75
## initial value 4975.946290
## iter 10 value 3357.656151
## iter 20 value 3228.829232
## iter 30 value 3213.935789
## iter 40 value 3211.597534
## iter 50 value 3210.869936
## iter 60 value 3210.262732
## iter 70 value 3209.150099
## final value 3209.098427
## converged
## # weights: 121
## initial value 5814.352037
## iter 10 value 3257.865768
## iter 20 value 3253.451983
## iter 30 value 3209.784239
## iter 40 value 3169.969167
## iter 50 value 2573.581385
## iter 60 value 2210.138645
## iter 70 value 2199.780383
## iter 80 value 2199.319486
## iter 90 value 2196.916544
## iter 100 value 2181.472778
## final value 2181.472778
## stopped after 100 iterations
## # weights: 167
## initial value 6512.357722
## iter 10 value 3275.396856
## iter 20 value 3242.563805
## iter 30 value 3230.791725
## iter 40 value 3074.020325
## iter 50 value 2434.700204
## iter 60 value 2261.382858
## iter 70 value 1913.411177
## iter 80 value 1737.357132
## iter 90 value 1641.177894
## iter 100 value 1438.838709
## final value 1438.838709
## stopped after 100 iterations
## # weights: 52
## initial value 7132.395340
## iter 10 value 3208.136385
## iter 20 value 3206.804131
## iter 30 value 3206.100160
## iter 40 value 3205.358468
## iter 50 value 3205.137744
## iter 60 value 2575.993049
## iter 70 value 2412.751543
## iter 80 value 2170.632916
## iter 90 value 2149.865792
## iter 100 value 2148.163641
## final value 2148.163641
## stopped after 100 iterations
## # weights: 75
## initial value 4978.498267
## iter 10 value 3207.652263
## iter 20 value 3206.321629
## iter 30 value 3205.006879
## iter 40 value 3200.808643
## iter 50 value 3129.480188
## iter 60 value 2727.268784
## iter 70 value 2151.470262
## iter 80 value 1673.133154
## iter 90 value 1158.132555
## iter 100 value 1098.012345
## final value 1098.012345
## stopped after 100 iterations
## # weights: 121
## initial value 5755.488742
## iter 10 value 3205.486778
## iter 20 value 3204.010982
## iter 30 value 3194.302822
## iter 40 value 2701.322264
## iter 50 value 2260.851053
## iter 60 value 2058.520417
## iter 70 value 2045.610560
## iter 80 value 1963.880942
## iter 90 value 1915.948403
## iter 100 value 1851.504210
## final value 1851.504210
## stopped after 100 iterations
## # weights: 167
## initial value 7009.434415
## iter 10 value 3248.046296
## iter 20 value 3209.011048
## iter 30 value 3203.300303
## iter 40 value 3203.153061
## iter 50 value 3202.978325
## final value 3202.949223
## converged
## # weights: 52
## initial value 5376.828592
## iter 10 value 3225.866314
## iter 20 value 3213.137427
## iter 30 value 3198.092421
## iter 40 value 3003.335997
## iter 50 value 2531.059993
## iter 60 value 2469.866515
## iter 70 value 2412.505864
## iter 80 value 2151.210013
## iter 90 value 1625.980984
## iter 100 value 1459.244503
## final value 1459.244503
## stopped after 100 iterations
## # weights: 75
## initial value 5787.321303
## iter 10 value 3211.168853
## iter 20 value 3208.122170
## iter 30 value 3207.738716
## iter 40 value 3206.078023
## final value 3206.065961
## converged
## # weights: 121
## initial value 4565.007791
## iter 10 value 3223.055506
## iter 20 value 3208.778015
## iter 30 value 3207.429969
## iter 40 value 3204.853135
## iter 50 value 3204.553150
## iter 60 value 3204.525540
## iter 70 value 3204.511848
## final value 3204.511135
## converged
## # weights: 167
## initial value 4975.723823
## iter 10 value 3264.337580
## iter 20 value 3220.391501
## iter 30 value 3204.662142
## iter 40 value 3204.502547
## iter 50 value 3204.288237
## iter 60 value 3203.959286
## iter 70 value 3203.627499
## iter 80 value 3165.937025
## iter 90 value 2312.757614
## iter 100 value 2261.152312
## final value 2261.152312
## stopped after 100 iterations
## # weights: 52
## initial value 5817.418350
## iter 10 value 3218.771371
## iter 20 value 3153.492815
## iter 30 value 2666.724047
## iter 40 value 2559.296979
## iter 50 value 2465.385156
## iter 60 value 2356.407787
## iter 70 value 2314.120129
## iter 80 value 2135.265566
## iter 90 value 1865.980224
## iter 100 value 1582.018866
## final value 1582.018866
## stopped after 100 iterations
## # weights: 75
## initial value 6925.396894
## iter 10 value 3286.827215
## iter 20 value 3259.239790
## iter 30 value 2536.659653
## iter 40 value 2164.543617
## iter 50 value 2066.840440
## iter 60 value 1996.407614
## iter 70 value 1946.039219
## iter 80 value 1909.763783
## iter 90 value 1779.259488
## iter 100 value 1607.939678
## final value 1607.939678
## stopped after 100 iterations
## # weights: 121
## initial value 6165.810721
## iter 10 value 3566.002920
## iter 20 value 3246.726347
## iter 30 value 3220.428777
## iter 40 value 3218.187981
## iter 50 value 3217.520696
## iter 60 value 3011.856135
## iter 70 value 2241.695507
## iter 80 value 2195.549969
## iter 90 value 1910.288934
## iter 100 value 1642.895088
## final value 1642.895088
## stopped after 100 iterations
## # weights: 167
## initial value 5094.142376
## iter 10 value 3210.514595
## iter 20 value 3210.058424
## iter 30 value 2639.038421
## iter 40 value 2292.989528
## iter 50 value 2221.219373
## iter 60 value 2100.222237
## iter 70 value 1370.245375
## iter 80 value 1280.786181
## iter 90 value 1240.034398
## iter 100 value 1201.875976
## final value 1201.875976
## stopped after 100 iterations
## # weights: 52
## initial value 4558.939341
## iter 10 value 3204.786183
## iter 20 value 3199.073017
## iter 30 value 2937.389981
## iter 40 value 2905.709589
## iter 50 value 2609.375588
## iter 60 value 2422.924594
## iter 70 value 2154.208386
## iter 80 value 2098.807759
## iter 90 value 2074.336754
## iter 100 value 2029.296401
## final value 2029.296401
## stopped after 100 iterations
## # weights: 75
## initial value 4663.912738
## iter 10 value 3202.357680
## iter 20 value 3200.592623
## iter 30 value 3199.516677
## iter 40 value 2901.861809
## iter 50 value 2629.615695
## iter 60 value 2426.119609
## iter 70 value 2282.281043
## iter 80 value 2101.800032
## iter 90 value 2080.215681
## iter 100 value 2055.259059
## final value 2055.259059
## stopped after 100 iterations
## # weights: 121
## initial value 6694.658852
## iter 10 value 3568.353442
## iter 20 value 3215.635290
## iter 30 value 3002.269470
## iter 40 value 2562.862081
## iter 50 value 2483.041267
## iter 60 value 2270.828245
## iter 70 value 2156.379372
## iter 80 value 1551.963280
## iter 90 value 1436.203595
## iter 100 value 1379.040178
## final value 1379.040178
## stopped after 100 iterations
## # weights: 167
## initial value 5884.920265
## iter 10 value 3263.011747
## iter 20 value 3242.468560
## iter 30 value 3026.184884
## iter 40 value 2400.748482
## iter 50 value 1869.598335
## iter 60 value 1297.107833
## iter 70 value 1224.136063
## iter 80 value 1188.718108
## iter 90 value 1149.356518
## iter 100 value 1138.044625
## final value 1138.044625
## stopped after 100 iterations
## # weights: 52
## initial value 5746.375140
## iter 10 value 3204.586556
## iter 20 value 3203.367424
## iter 30 value 2835.532229
## iter 40 value 2132.790297
## iter 50 value 2090.760053
## iter 60 value 2005.730023
## iter 70 value 1910.977213
## iter 80 value 1599.748714
## iter 90 value 1256.125717
## iter 100 value 1064.163094
## final value 1064.163094
## stopped after 100 iterations
## # weights: 75
## initial value 4986.270698
## iter 10 value 3222.914866
## iter 20 value 3212.079649
## iter 30 value 3180.213478
## iter 40 value 2973.227614
## iter 50 value 2783.463514
## iter 60 value 2369.660407
## iter 70 value 2339.900527
## iter 80 value 2309.213379
## iter 90 value 1548.301725
## iter 100 value 1323.077052
## final value 1323.077052
## stopped after 100 iterations
## # weights: 121
## initial value 6547.087385
## iter 10 value 3214.011311
## iter 20 value 3210.142264
## iter 30 value 2498.570812
## iter 40 value 2297.796104
## iter 50 value 2263.093903
## iter 60 value 2097.329159
## iter 70 value 2046.668950
## iter 80 value 1992.017942
## iter 90 value 1930.505396
## iter 100 value 1792.799062
## final value 1792.799062
## stopped after 100 iterations
## # weights: 167
## initial value 4684.515706
## iter 10 value 3269.861268
## iter 20 value 3206.278317
## iter 30 value 3199.944293
## iter 40 value 3199.175602
## iter 50 value 3199.072853
## iter 60 value 3198.841774
## iter 70 value 3198.587109
## iter 80 value 3198.319484
## iter 90 value 3178.000195
## iter 100 value 2791.155216
## final value 2791.155216
## stopped after 100 iterations
## # weights: 52
## initial value 6515.484485
## iter 10 value 3239.279085
## iter 20 value 2794.431933
## iter 30 value 2668.544819
## iter 40 value 2490.410065
## iter 50 value 1976.884036
## iter 60 value 1488.038850
## iter 70 value 1269.840333
## iter 80 value 1163.603500
## iter 90 value 1103.683873
## iter 100 value 1052.674018
## final value 1052.674018
## stopped after 100 iterations
## # weights: 75
## initial value 5286.435465
## iter 10 value 3240.794527
## iter 20 value 3206.244899
## iter 30 value 3203.393345
## iter 40 value 3020.631234
## iter 50 value 2793.961578
## iter 60 value 2537.222158
## iter 70 value 2222.256966
## iter 80 value 1928.377165
## iter 90 value 1564.694548
## iter 100 value 1354.159807
## final value 1354.159807
## stopped after 100 iterations
## # weights: 121
## initial value 4907.230200
## iter 10 value 3209.074842
## iter 20 value 3206.353658
## iter 30 value 3202.119353
## iter 40 value 3201.595631
## iter 50 value 3173.860094
## iter 60 value 2935.861469
## iter 70 value 2607.174746
## iter 80 value 2107.255502
## iter 90 value 1877.865394
## iter 100 value 1444.543010
## final value 1444.543010
## stopped after 100 iterations
## # weights: 167
## initial value 4681.160404
## iter 10 value 3238.217264
## iter 20 value 3214.840187
## iter 30 value 3203.448259
## iter 40 value 3129.524036
## iter 50 value 2710.621977
## iter 60 value 2244.415136
## iter 70 value 2213.319733
## iter 80 value 2207.217669
## iter 90 value 2102.199359
## iter 100 value 1252.378460
## final value 1252.378460
## stopped after 100 iterations
## # weights: 167
## initial value 8574.336580
## iter 10 value 5078.857550
## iter 20 value 4812.122155
## iter 30 value 4811.215353
## iter 40 value 4745.905009
## iter 50 value 3725.694534
## iter 60 value 3454.730277
## iter 70 value 3309.245039
## iter 80 value 2975.425516
## iter 90 value 2433.686206
## iter 100 value 2068.920154
## final value 2068.920154
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n3_NN1Fit0
## Neural Network
##
## 4149 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 2766, 2767, 2765
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.7334305 0.5791695
## 2 0.5 0.7160733 0.5460594
## 2 0.7 0.8365712 0.7511677
## 3 0.3 0.7797566 0.6554878
## 3 0.5 0.7193306 0.5511962
## 3 0.7 0.6934120 0.5119131
## 5 0.3 0.7541269 0.6158524
## 5 0.5 0.5111167 0.2104419
## 5 0.7 0.7929564 0.6754986
## 7 0.3 0.7075162 0.5327891
## 7 0.5 0.7083867 0.5340708
## 7 0.7 0.8756300 0.8108256
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 7 and decay = 0.7.
DryBean_TDA_KDE_5.50.5_n3_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.8791606 0.8159733 Fold2
## 2 0.8575560 0.7837989 Fold1
## 3 0.8901734 0.8327046 Fold3
nb_tda_kde_5.50.5_n3_nn1_fit_re<-DryBean_TDA_KDE_5.50.5_n3_NN1Fit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n3_NN1Fit0)
## a 16-7-6 network with 167 weights
## options were - softmax modelling decay=0.7
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.02 0.62 4.81 9.54 -2.75 0.11 0.08 -0.77 3.00 0.08
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.02 0.04 -0.01 0.00 0.00 -0.03 0.03
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 -0.01 0.00 0.00 0.00 0.00 0.00 -0.01 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 -1.55 -0.10 4.25 -4.86 0.05 0.03 1.55 -1.09 -0.01
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## -0.01 -0.01 -0.02 0.00 0.00 -0.03 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 -0.05 0.00 0.00 0.00 0.00 0.00 -0.05 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h6 i1->h6 i2->h6 i3->h6 i4->h6 i5->h6 i6->h6 i7->h6 i8->h6 i9->h6
## -0.01 -0.24 -0.15 0.58 -3.06 -0.01 0.10 0.25 0.40 -0.04
## i10->h6 i11->h6 i12->h6 i13->h6 i14->h6 i15->h6 i16->h6
## -0.01 -0.03 -0.04 0.00 0.00 -0.07 -0.02
## b->h7 i1->h7 i2->h7 i3->h7 i4->h7 i5->h7 i6->h7 i7->h7 i8->h7 i9->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h7 i11->h7 i12->h7 i13->h7 i14->h7 i15->h7 i16->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1 h6->o1 h7->o1
## -0.96 -0.82 1.41 -0.97 -0.34 -0.06 -0.06 0.01
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2 h6->o2 h7->o2
## -0.72 -0.43 -1.74 -0.73 -0.30 -0.44 -0.16 0.02
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3 h6->o3 h7->o3
## 0.11 3.70 0.02 0.12 0.66 0.15 -0.32 0.00
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4 h6->o4 h7->o4
## -0.98 -0.21 -0.26 -0.98 1.50 1.43 0.73 0.00
## b->o5 h1->o5 h2->o5 h3->o5 h4->o5 h5->o5 h6->o5 h7->o5
## 2.01 -1.80 0.44 2.02 -1.49 -0.39 -2.14 -0.01
## b->o6 h1->o6 h2->o6 h3->o6 h4->o6 h5->o6 h6->o6 h7->o6
## 0.54 -0.43 0.11 0.54 -0.02 -0.69 1.94 -0.01
#vip(DryBean_TDA_KDE_5.50.5_n3_NN1Fit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n3_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n3_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n3_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n3_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n3_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 945 105 62 94
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 37 0 485 13
## SIRA 396 156 489 81 473 61 683
##
## Overall Statistics
##
## Accuracy : 0.5179
## 95% CI : (0.5024, 0.5333)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3916
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.8890
## Specificity 1.00000 1.00000 1.0000 0.9135
## Pos Pred Value NaN NaN NaN 0.7836
## Neg Pred Value 0.90294 0.96176 0.8801 0.9589
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2316
## Detection Prevalence 0.00000 0.00000 0.0000 0.2956
## Balanced Accuracy 0.50000 0.50000 0.5000 0.9012
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.7977 0.8646
## Specificity 1.0000 0.9856 0.4967
## Pos Pred Value NaN 0.9065 0.2920
## Neg Pred Value 0.8583 0.9653 0.9385
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1189 0.1674
## Detection Prevalence 0.0000 0.1311 0.5733
## Balanced Accuracy 0.5000 0.8916 0.6806
nb_tda_kde_5.50.5_n3_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 945 105 62 94
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 37 0 485 13
## SIRA 396 156 489 81 473 61 683
##
## Overall Statistics
##
## Accuracy : 0.5179
## 95% CI : (0.5024, 0.5333)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3916
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.8890
## Specificity 1.00000 1.00000 1.0000 0.9135
## Pos Pred Value NaN NaN NaN 0.7836
## Neg Pred Value 0.90294 0.96176 0.8801 0.9589
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2316
## Detection Prevalence 0.00000 0.00000 0.0000 0.2956
## Balanced Accuracy 0.50000 0.50000 0.5000 0.9012
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.7977 0.8646
## Specificity 1.0000 0.9856 0.4967
## Pos Pred Value NaN 0.9065 0.2920
## Neg Pred Value 0.8583 0.9653 0.9385
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1189 0.1674
## Detection Prevalence 0.0000 0.1311 0.5733
## Balanced Accuracy 0.5000 0.8916 0.6806
nb_tda_kde_5.50.5_n3_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.178922e-01 3.916187e-01 5.024290e-01 5.333296e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 1.709603e-267 NaN
nb_tda_kde_5.50.5_n3_db_nn1_cf0_ov_acc<-nb_tda_kde_5.50.5_n3_db_nn1_cf0$overall[1]
nb_tda_kde_5.50.5_n3_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.8889934 0.9134902 0.7835821 0.9589422 0.7835821
## Class: HOROZ 0.0000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.7976974 0.9855991 0.9065421 0.9653032 0.9065421
## Class: SIRA 0.8645570 0.4966565 0.2920051 0.9385411 0.2920051
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.0000000
## Class: BOMBAY 0.0000000 NA 0.03823529 0.0000000
## Class: CALI 0.0000000 NA 0.11985294 0.0000000
## Class: DERMASON 0.8889934 0.8329661 0.26053922 0.2316176
## Class: HOROZ 0.0000000 NA 0.14166667 0.0000000
## Class: SEKER 0.7976974 0.8486439 0.14901961 0.1188725
## Class: SIRA 0.8645570 0.4365612 0.19362745 0.1674020
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000 0.5000000
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.0000000 0.5000000
## Class: DERMASON 0.2955882 0.9012418
## Class: HOROZ 0.0000000 0.5000000
## Class: SEKER 0.1311275 0.8916482
## Class: SIRA 0.5732843 0.6806067
nb_tda_kde_5.50.5_n3_db_nn1_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n3_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold<-(db_nn1_fit_re - nb_tda_kde_5.50.5_n3_nn1_fit_re)
diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold
## Accuracy
## 1 -0.1735561
## 2 -0.3243486
## 3 -0.3228355
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n3_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n3_3_fold_odds.left<-bst_tda_kde_5.50.5_nn1.n3_3_fold$probLeft/bst_tda_kde_5.50.5_nn1.n3_3_fold$probRight
bst_tda_kde_5.50.5_nn1.n3_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n3_3_fold
## $winLeft
## [1] 0.9903333
##
## $winRope
## [1] 0.009666667
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n3_3_fold
## $left
## [1] 0.9775979
##
## $rope
## [1] 0.002872362
##
## $right
## [1] 0.01952977
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold))
#bf_tda_kde_5.50.5_nn1.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n3_3_fold)
## t = -5.4701, df = 2, p-value = 0.03183
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.48877244 -0.05838768
## sample estimates:
## mean of x
## -0.2735801
### Test set diff
diff_drybean_tda_kde_5.50.5_nn1.n3_test<-(db_nn1_cf_ov_acc - nb_tda_kde_5.50.5_n3_db_nn1_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nn1.n3_test
## Accuracy
## 0.09191176
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n3_test),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n3_test_odds.left<-bst_tda_kde_5.50.5_nn1.n3_test$probLeft/bst_tda_kde_5.50.5_nn1.n3_test$probRight
bst_tda_kde_5.50.5_nn1.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n3_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1586667
##
## $winRight
## [1] 0.8413333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n3_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nn1.n3_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n3_test)) #bf_tda_pca_5.50.5_nn1.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n3_test))
##Node4
#Neural Network 1
DryBean_TDA_KDE_5.50.5_n4_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n4.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 46
## initial value 1711.865013
## iter 10 value 1410.151303
## iter 20 value 1312.981656
## iter 30 value 869.937500
## iter 40 value 706.326657
## iter 50 value 665.232862
## iter 60 value 633.023238
## iter 70 value 620.280055
## iter 80 value 609.643320
## iter 90 value 594.173366
## final value 594.148556
## converged
## # weights: 67
## initial value 2177.975098
## iter 10 value 1405.900831
## iter 20 value 1363.901707
## iter 30 value 1215.539384
## iter 40 value 1092.582311
## iter 50 value 847.309124
## iter 60 value 736.642930
## iter 70 value 714.830862
## iter 80 value 667.066295
## iter 90 value 604.653230
## iter 100 value 576.018864
## final value 576.018864
## stopped after 100 iterations
## # weights: 109
## initial value 1628.555046
## iter 10 value 1405.736635
## iter 20 value 1404.015223
## iter 30 value 1403.890476
## iter 40 value 1392.305397
## iter 50 value 1349.041713
## iter 60 value 1108.131072
## iter 70 value 924.034051
## iter 80 value 724.003613
## iter 90 value 646.316131
## iter 100 value 613.900550
## final value 613.900550
## stopped after 100 iterations
## # weights: 151
## initial value 2461.616664
## iter 10 value 1405.997035
## iter 20 value 1403.882363
## final value 1403.793911
## converged
## # weights: 46
## initial value 2097.721965
## iter 10 value 1415.041269
## iter 20 value 1406.934972
## iter 30 value 1406.648133
## iter 40 value 1406.641446
## iter 50 value 1375.329702
## iter 60 value 1332.386260
## iter 70 value 1267.143124
## iter 80 value 957.454435
## iter 90 value 834.157511
## iter 100 value 804.807143
## final value 804.807143
## stopped after 100 iterations
## # weights: 67
## initial value 1777.670696
## iter 10 value 1408.235141
## iter 20 value 1406.550345
## iter 30 value 1196.313444
## iter 40 value 912.550545
## iter 50 value 829.365510
## iter 60 value 754.469162
## iter 70 value 672.805206
## iter 80 value 618.225280
## iter 90 value 608.862322
## iter 100 value 593.904818
## final value 593.904818
## stopped after 100 iterations
## # weights: 109
## initial value 1970.643224
## iter 10 value 1416.274064
## iter 20 value 1405.570549
## iter 30 value 1404.960786
## iter 40 value 1404.554677
## iter 50 value 1404.350181
## iter 60 value 1404.345722
## iter 70 value 1404.335666
## final value 1404.335223
## converged
## # weights: 151
## initial value 2458.500221
## iter 10 value 1414.928662
## iter 20 value 1407.726928
## iter 30 value 1386.479127
## iter 40 value 1131.433144
## iter 50 value 842.381856
## iter 60 value 816.174918
## iter 70 value 790.417366
## iter 80 value 774.460270
## iter 90 value 763.561740
## iter 100 value 762.971086
## final value 762.971086
## stopped after 100 iterations
## # weights: 46
## initial value 1563.689212
## iter 10 value 1408.606751
## iter 20 value 1407.959539
## iter 30 value 1407.951090
## iter 40 value 1407.898128
## iter 50 value 1407.229962
## iter 60 value 1391.760584
## iter 70 value 1281.471173
## iter 80 value 1208.229396
## iter 90 value 1174.912863
## iter 100 value 1151.286079
## final value 1151.286079
## stopped after 100 iterations
## # weights: 67
## initial value 1779.838415
## iter 10 value 1406.429648
## iter 20 value 1406.420484
## iter 30 value 1406.016224
## iter 40 value 1405.634479
## final value 1405.626836
## converged
## # weights: 109
## initial value 1540.035371
## iter 10 value 1413.830765
## iter 20 value 1408.004695
## iter 30 value 1406.614031
## iter 40 value 1404.526111
## iter 50 value 1142.439615
## iter 60 value 922.823426
## iter 70 value 867.452444
## iter 80 value 848.823899
## iter 90 value 811.911889
## iter 100 value 799.144129
## final value 799.144129
## stopped after 100 iterations
## # weights: 151
## initial value 1872.859619
## iter 10 value 1405.740191
## iter 20 value 1405.157420
## iter 30 value 1404.791395
## iter 40 value 1404.676988
## iter 50 value 1250.429186
## iter 60 value 1128.748171
## iter 70 value 951.310440
## iter 80 value 817.717083
## iter 90 value 796.842058
## iter 100 value 779.449223
## final value 779.449223
## stopped after 100 iterations
## # weights: 46
## initial value 2378.294149
## iter 10 value 1405.397029
## iter 20 value 1405.273548
## iter 30 value 1249.869510
## iter 40 value 1115.712458
## iter 50 value 887.923343
## iter 60 value 833.673131
## iter 70 value 806.948572
## iter 80 value 796.964009
## iter 90 value 794.140611
## iter 100 value 766.246715
## final value 766.246715
## stopped after 100 iterations
## # weights: 67
## initial value 2008.976245
## iter 10 value 1405.360307
## iter 20 value 1404.591013
## iter 30 value 1404.576393
## final value 1404.576293
## converged
## # weights: 109
## initial value 1753.692532
## iter 10 value 1410.997445
## iter 20 value 1405.003287
## iter 30 value 1403.754626
## iter 40 value 1323.602037
## iter 50 value 1172.956433
## iter 60 value 940.144850
## iter 70 value 802.622142
## iter 80 value 774.409781
## iter 90 value 770.521520
## iter 100 value 758.275708
## final value 758.275708
## stopped after 100 iterations
## # weights: 151
## initial value 1510.428290
## iter 10 value 1403.864307
## iter 20 value 1403.850731
## final value 1403.850661
## converged
## # weights: 46
## initial value 1945.851951
## iter 10 value 1410.100335
## iter 20 value 1406.729296
## iter 30 value 1374.215401
## iter 40 value 1096.579929
## iter 50 value 868.034401
## iter 60 value 799.629437
## iter 70 value 755.269992
## iter 80 value 704.442267
## iter 90 value 685.109483
## iter 100 value 655.262540
## final value 655.262540
## stopped after 100 iterations
## # weights: 67
## initial value 2173.906205
## iter 10 value 1414.995832
## iter 20 value 1406.896726
## iter 30 value 1406.645732
## iter 40 value 1406.629848
## iter 50 value 1405.421837
## iter 60 value 1405.194250
## iter 70 value 1372.269529
## iter 80 value 860.619201
## iter 90 value 818.846319
## iter 100 value 784.657503
## final value 784.657503
## stopped after 100 iterations
## # weights: 109
## initial value 1947.370857
## iter 10 value 1413.202404
## iter 20 value 1407.169849
## iter 30 value 1405.209291
## iter 40 value 1405.082090
## iter 50 value 1404.680813
## iter 60 value 1362.328046
## iter 70 value 1324.166410
## iter 80 value 1061.408923
## iter 90 value 902.608538
## iter 100 value 824.665734
## final value 824.665734
## stopped after 100 iterations
## # weights: 151
## initial value 2414.734218
## iter 10 value 1314.489082
## iter 20 value 1116.208427
## iter 30 value 1050.060915
## iter 40 value 920.167352
## iter 50 value 846.685294
## iter 60 value 815.477982
## iter 70 value 728.887118
## iter 80 value 677.116142
## iter 90 value 645.090689
## iter 100 value 621.999209
## final value 621.999209
## stopped after 100 iterations
## # weights: 46
## initial value 2506.161810
## iter 10 value 1406.422707
## iter 20 value 1406.420649
## final value 1406.420601
## converged
## # weights: 67
## initial value 1917.352674
## iter 10 value 1411.877725
## iter 20 value 1409.787337
## iter 30 value 1337.813977
## iter 40 value 961.094897
## iter 50 value 837.558020
## iter 60 value 826.187459
## iter 70 value 801.145019
## iter 80 value 785.229234
## iter 90 value 771.778944
## final value 771.745950
## converged
## # weights: 109
## initial value 1949.450429
## iter 10 value 1415.715749
## iter 20 value 1409.597042
## iter 30 value 1406.261196
## iter 40 value 1295.651499
## iter 50 value 1125.940204
## iter 60 value 859.008726
## iter 70 value 832.522930
## iter 80 value 816.494127
## iter 90 value 746.843918
## iter 100 value 715.664424
## final value 715.664424
## stopped after 100 iterations
## # weights: 151
## initial value 1523.585544
## iter 10 value 1410.122590
## iter 20 value 1404.942577
## iter 30 value 1404.811209
## iter 40 value 1404.589926
## iter 50 value 1404.576550
## iter 50 value 1404.576540
## iter 60 value 1404.571593
## iter 70 value 1404.421722
## iter 80 value 1404.395464
## final value 1404.395196
## converged
## # weights: 46
## initial value 2043.209824
## iter 10 value 1409.979127
## iter 20 value 1409.943730
## iter 30 value 1408.683814
## iter 40 value 1259.522159
## iter 50 value 1252.295744
## iter 60 value 1079.747052
## iter 70 value 835.436834
## iter 80 value 801.319515
## iter 90 value 781.486227
## iter 100 value 755.631657
## final value 755.631657
## stopped after 100 iterations
## # weights: 67
## initial value 1835.760337
## iter 10 value 1409.550851
## iter 20 value 1409.184363
## final value 1408.990786
## converged
## # weights: 109
## initial value 1954.385964
## iter 10 value 1413.922047
## iter 20 value 1407.203945
## iter 30 value 1177.112264
## iter 40 value 992.144778
## iter 50 value 976.006050
## iter 60 value 867.733247
## iter 70 value 825.283859
## iter 80 value 717.283089
## iter 90 value 665.663901
## iter 100 value 662.664976
## final value 662.664976
## stopped after 100 iterations
## # weights: 151
## initial value 2821.412506
## iter 10 value 1419.093101
## final value 1408.982681
## converged
## # weights: 46
## initial value 2416.380019
## iter 10 value 1419.028012
## iter 20 value 1411.496318
## iter 30 value 1411.250820
## final value 1411.247167
## converged
## # weights: 67
## initial value 1715.597492
## iter 10 value 1417.368767
## iter 20 value 1411.417113
## iter 30 value 1411.248404
## iter 40 value 1409.858912
## iter 50 value 1409.645289
## final value 1409.644958
## converged
## # weights: 109
## initial value 2312.888515
## iter 10 value 1414.336647
## iter 20 value 1409.407812
## iter 30 value 1409.319823
## iter 40 value 1409.299754
## iter 50 value 1409.163957
## iter 60 value 1283.076438
## iter 70 value 1234.110419
## iter 80 value 999.146136
## iter 90 value 882.054324
## iter 100 value 794.244523
## final value 794.244523
## stopped after 100 iterations
## # weights: 151
## initial value 2803.998688
## iter 10 value 1422.983779
## iter 20 value 1356.092931
## iter 30 value 1193.561394
## iter 40 value 831.792869
## iter 50 value 783.063072
## iter 60 value 764.528941
## iter 70 value 738.052647
## iter 80 value 710.889011
## iter 90 value 709.124307
## iter 100 value 704.100869
## final value 704.100869
## stopped after 100 iterations
## # weights: 46
## initial value 2258.495038
## iter 10 value 1431.015938
## iter 20 value 1419.949993
## iter 30 value 1312.413572
## iter 40 value 1228.200927
## iter 50 value 1044.392163
## iter 60 value 850.450986
## iter 70 value 818.337974
## iter 80 value 791.570314
## iter 90 value 784.207982
## iter 100 value 763.710726
## final value 763.710726
## stopped after 100 iterations
## # weights: 67
## initial value 2702.192678
## iter 10 value 1423.340425
## iter 20 value 1412.863839
## iter 30 value 1412.479173
## iter 40 value 1412.436124
## iter 50 value 1390.835481
## iter 60 value 1368.240813
## iter 70 value 1190.663287
## iter 80 value 1121.465931
## iter 90 value 1083.907508
## iter 100 value 827.274825
## final value 827.274825
## stopped after 100 iterations
## # weights: 109
## initial value 2597.462562
## iter 10 value 1423.744886
## iter 20 value 1401.366982
## iter 30 value 1258.217610
## iter 40 value 1155.094128
## iter 50 value 980.708767
## iter 60 value 871.975542
## iter 70 value 822.770209
## iter 80 value 796.233453
## iter 90 value 783.479191
## iter 100 value 781.569369
## final value 781.569369
## stopped after 100 iterations
## # weights: 151
## initial value 2877.548634
## iter 10 value 1421.115682
## iter 20 value 1418.811898
## iter 30 value 1418.434061
## iter 40 value 1352.462071
## iter 50 value 1191.303792
## iter 60 value 852.399171
## iter 70 value 808.933550
## iter 80 value 804.427499
## iter 90 value 793.229162
## iter 100 value 776.909736
## final value 776.909736
## stopped after 100 iterations
## # weights: 109
## initial value 4567.744758
## iter 10 value 2108.539681
## iter 20 value 2107.861885
## final value 2107.847391
## converged
DryBean_TDA_KDE_5.50.5_n4_NN1Fit0
## Neural Network
##
## 2024 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1349, 1349, 1350
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.7746947 0.6027428
## 2 0.5 0.6876638 0.4069571
## 2 0.7 0.6176591 0.2870128
## 3 0.3 0.6111206 0.2216656
## 3 0.5 0.6935898 0.4129982
## 3 0.7 0.6779045 0.3723171
## 5 0.3 0.7875569 0.6352066
## 5 0.5 0.6714760 0.3597457
## 5 0.7 0.7771653 0.6059644
## 7 0.3 0.5197626 0.0000000
## 7 0.5 0.7830904 0.6160797
## 7 0.7 0.6769161 0.3713985
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 5 and decay = 0.3.
DryBean_TDA_KDE_5.50.5_n4_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.8059259 0.6830327 Fold1
## 2 0.8026706 0.6711532 Fold3
## 3 0.7540741 0.5514340 Fold2
nb_tda_kde_5.50.5_n4_nn1_fit_re<-DryBean_TDA_KDE_5.50.5_n4_NN1Fit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n4_NN1Fit0)
## a 16-5-4 network with 109 weights
## options were - softmax modelling decay=0.3
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.08 0.00 0.00
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1
## 0.27 0.27 0.27 0.27 0.27 0.27
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2
## -0.54 -0.54 -0.54 -0.54 -0.54 -0.54
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3
## 0.16 0.16 0.16 0.16 0.16 0.16
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4
## 0.11 0.11 0.11 0.11 0.11 0.11
#vip(DryBean_TDA_KDE_5.50.5_n4_NN1Fit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n4_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n4_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n4_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n4_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n4_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 396 156 489 1063 578 608 790
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2605
## 95% CI : (0.2471, 0.2743)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.506
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 1.0000
## Specificity 1.00000 1.00000 1.0000 0.0000
## Pos Pred Value NaN NaN NaN 0.2605
## Neg Pred Value 0.90294 0.96176 0.8801 NaN
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2605
## Detection Prevalence 0.00000 0.00000 0.0000 1.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
nb_tda_kde_5.50.5_n4_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 396 156 489 1063 578 608 790
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.2605
## 95% CI : (0.2471, 0.2743)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 0.506
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 1.0000
## Specificity 1.00000 1.00000 1.0000 0.0000
## Pos Pred Value NaN NaN NaN 0.2605
## Neg Pred Value 0.90294 0.96176 0.8801 NaN
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2605
## Detection Prevalence 0.00000 0.00000 0.0000 1.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
nb_tda_kde_5.50.5_n4_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.2605392 0.0000000 0.2471237 0.2742985 0.2605392
## AccuracyPValue McnemarPValue
## 0.5059787 NaN
nb_tda_kde_5.50.5_n4_db_nn1_cf0_ov_acc<-nb_tda_kde_5.50.5_n4_db_nn1_cf0$overall[1]
nb_tda_kde_5.50.5_n4_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0 1 NaN 0.9029412 NA
## Class: BOMBAY 0 1 NaN 0.9617647 NA
## Class: CALI 0 1 NaN 0.8801471 NA
## Class: DERMASON 1 0 0.2605392 NaN 0.2605392
## Class: HOROZ 0 1 NaN 0.8583333 NA
## Class: SEKER 0 1 NaN 0.8509804 NA
## Class: SIRA 0 1 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate Detection Prevalence
## Class: BARBUNYA 0 NA 0.09705882 0.0000000 0
## Class: BOMBAY 0 NA 0.03823529 0.0000000 0
## Class: CALI 0 NA 0.11985294 0.0000000 0
## Class: DERMASON 1 0.4133774 0.26053922 0.2605392 1
## Class: HOROZ 0 NA 0.14166667 0.0000000 0
## Class: SEKER 0 NA 0.14901961 0.0000000 0
## Class: SIRA 0 NA 0.19362745 0.0000000 0
## Balanced Accuracy
## Class: BARBUNYA 0.5
## Class: BOMBAY 0.5
## Class: CALI 0.5
## Class: DERMASON 0.5
## Class: HOROZ 0.5
## Class: SEKER 0.5
## Class: SIRA 0.5
nb_tda_kde_5.50.5_n4_db_nn1_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n4_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold<-(db_nn1_fit_re - nb_tda_kde_5.50.5_n4_nn1_fit_re)
diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold
## Accuracy
## 1 -0.1003214
## 2 -0.2694632
## 3 -0.1867361
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n4_3_fold_odds.left<-bst_tda_kde_5.50.5_nn1.n4_3_fold$probLeft/bst_tda_kde_5.50.5_nn1.n4_3_fold$probRight
bst_tda_kde_5.50.5_nn1.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n4_3_fold
## $winLeft
## [1] 0.9910667
##
## $winRope
## [1] 0.008933333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n4_3_fold
## $left
## [1] 0.9552178
##
## $rope
## [1] 0.007754273
##
## $right
## [1] 0.03702797
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold))
#bf_tda_kde_5.50.5_nn1.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n4_3_fold)
## t = -3.799, df = 2, p-value = 0.06283
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.3956093 0.0245955
## sample estimates:
## mean of x
## -0.1855069
### Test set diff
diff_drybean_tda_kde_5.50.5_nn1.n4_test<-(db_nn1_cf_ov_acc - nb_tda_kde_5.50.5_n4_db_nn1_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nn1.n4_test
## Accuracy
## 0.3492647
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n4_test_odds.left<-bst_tda_kde_5.50.5_nn1.n4_test$probLeft/bst_tda_kde_5.50.5_nn1.n4_test$probRight
bst_tda_kde_5.50.5_nn1.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1568333
##
## $winRight
## [1] 0.8431667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nn1.n4_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test)) #bf_tda_pca_5.50.5_nn1.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test))
##Node5
#Neural Network 1
DryBean_TDA_KDE_5.50.5_n5_NN1Fit0 <- train(as.factor(Class) ~ ., data = tda.m_kde_dry_bean_dataset_5.50.5.n5.vec,
Importance = T,
method = 'nnet',
trControl = fitControl,
tuneGrid = nn1Grid,
metric='Accuracy')
## # weights: 46
## initial value 1083.296816
## iter 10 value 661.912163
## iter 20 value 659.703185
## iter 30 value 659.692990
## final value 659.692912
## converged
## # weights: 67
## initial value 1146.084052
## iter 10 value 662.240869
## iter 20 value 661.768047
## iter 30 value 648.269821
## iter 40 value 542.499740
## iter 50 value 458.071769
## iter 60 value 453.643266
## iter 70 value 448.041110
## iter 80 value 428.229776
## iter 90 value 414.793091
## iter 100 value 393.881233
## final value 393.881233
## stopped after 100 iterations
## # weights: 109
## initial value 1966.706792
## iter 10 value 661.431775
## iter 20 value 658.276992
## iter 30 value 658.056934
## iter 40 value 654.202860
## iter 50 value 653.137195
## iter 60 value 555.507407
## iter 70 value 514.680066
## iter 80 value 492.442697
## iter 90 value 469.866079
## iter 100 value 456.829561
## final value 456.829561
## stopped after 100 iterations
## # weights: 151
## initial value 1136.840272
## iter 10 value 666.992771
## iter 20 value 658.655091
## iter 30 value 658.580966
## iter 40 value 652.910083
## iter 50 value 638.994397
## iter 60 value 638.255321
## iter 70 value 627.910576
## iter 80 value 539.451702
## iter 90 value 488.393781
## iter 100 value 445.926432
## final value 445.926432
## stopped after 100 iterations
## # weights: 46
## initial value 768.596848
## iter 10 value 661.889019
## iter 20 value 661.739697
## iter 30 value 661.346234
## iter 40 value 564.722647
## iter 50 value 502.938420
## iter 60 value 465.920408
## iter 70 value 452.267574
## iter 80 value 417.144812
## iter 90 value 406.071972
## iter 100 value 405.976515
## final value 405.976515
## stopped after 100 iterations
## # weights: 67
## initial value 942.580317
## iter 10 value 662.295607
## iter 20 value 658.018628
## iter 30 value 602.492044
## iter 40 value 546.304489
## iter 50 value 448.544339
## iter 60 value 436.477101
## iter 70 value 434.923842
## iter 80 value 434.889023
## iter 90 value 434.873973
## iter 100 value 433.594921
## final value 433.594921
## stopped after 100 iterations
## # weights: 109
## initial value 1259.519784
## iter 10 value 681.933409
## iter 20 value 659.537516
## iter 30 value 554.932248
## iter 40 value 512.844560
## iter 50 value 480.426645
## iter 60 value 452.631911
## iter 70 value 426.178520
## iter 80 value 421.129940
## iter 90 value 412.956881
## iter 100 value 408.866546
## final value 408.866546
## stopped after 100 iterations
## # weights: 151
## initial value 1090.278037
## iter 10 value 663.554003
## iter 20 value 660.647624
## iter 30 value 659.587640
## iter 40 value 656.787749
## iter 50 value 615.408076
## iter 60 value 583.054884
## iter 70 value 544.170361
## iter 80 value 532.996225
## iter 90 value 479.269623
## iter 100 value 453.105981
## final value 453.105981
## stopped after 100 iterations
## # weights: 46
## initial value 1048.359518
## iter 10 value 672.063095
## iter 20 value 667.010432
## iter 30 value 631.704457
## iter 40 value 493.367634
## iter 50 value 450.115296
## iter 60 value 442.270164
## final value 442.268447
## converged
## # weights: 67
## initial value 1078.499825
## iter 10 value 660.405445
## iter 20 value 660.096397
## final value 660.093951
## converged
## # weights: 109
## initial value 726.426094
## iter 10 value 664.468963
## iter 20 value 659.736790
## iter 30 value 659.535082
## iter 40 value 659.498221
## iter 50 value 659.204863
## iter 60 value 659.165219
## iter 70 value 659.142292
## final value 659.137815
## converged
## # weights: 151
## initial value 689.031583
## iter 10 value 661.244627
## iter 20 value 660.109532
## iter 30 value 660.094244
## final value 660.094184
## converged
## # weights: 46
## initial value 1101.256179
## iter 10 value 669.206853
## iter 20 value 668.582319
## iter 30 value 665.603720
## iter 40 value 613.097167
## iter 50 value 486.033338
## iter 60 value 469.502237
## iter 70 value 438.908871
## iter 80 value 426.661638
## iter 90 value 403.653933
## iter 100 value 399.225500
## final value 399.225500
## stopped after 100 iterations
## # weights: 67
## initial value 1058.043949
## iter 10 value 667.487504
## iter 20 value 666.945292
## iter 30 value 664.398733
## iter 40 value 569.432910
## iter 50 value 479.339373
## iter 60 value 470.449141
## iter 70 value 457.710397
## iter 80 value 448.203644
## iter 90 value 440.389560
## iter 100 value 432.793379
## final value 432.793379
## stopped after 100 iterations
## # weights: 109
## initial value 859.130258
## iter 10 value 666.630789
## iter 20 value 665.550652
## iter 30 value 665.506270
## iter 40 value 588.918336
## iter 50 value 499.922795
## iter 60 value 470.011610
## iter 70 value 453.394337
## iter 80 value 440.413304
## iter 90 value 438.462205
## iter 100 value 432.695209
## final value 432.695209
## stopped after 100 iterations
## # weights: 151
## initial value 1279.730501
## iter 10 value 744.135382
## iter 20 value 633.939324
## iter 30 value 624.292255
## iter 40 value 594.955618
## iter 50 value 549.804395
## iter 60 value 490.812239
## iter 70 value 444.389754
## iter 80 value 415.904658
## iter 90 value 405.740346
## iter 100 value 395.552599
## final value 395.552599
## stopped after 100 iterations
## # weights: 46
## initial value 930.206589
## iter 10 value 669.035284
## iter 20 value 667.114359
## iter 30 value 667.072866
## final value 667.072709
## converged
## # weights: 67
## initial value 1115.948881
## iter 10 value 682.904581
## iter 20 value 675.837263
## iter 30 value 667.127851
## iter 40 value 666.732975
## iter 50 value 666.582811
## iter 60 value 660.108424
## iter 70 value 650.068772
## iter 80 value 564.957284
## iter 90 value 458.072837
## iter 100 value 436.975679
## final value 436.975679
## stopped after 100 iterations
## # weights: 109
## initial value 919.830648
## iter 10 value 672.260257
## iter 20 value 666.256759
## iter 30 value 660.539098
## iter 40 value 624.441703
## iter 50 value 507.602535
## iter 60 value 464.442259
## iter 70 value 455.928778
## iter 80 value 450.389771
## iter 90 value 445.554633
## iter 100 value 440.479437
## final value 440.479437
## stopped after 100 iterations
## # weights: 151
## initial value 802.791351
## iter 10 value 668.709498
## iter 20 value 667.177925
## iter 30 value 666.995905
## iter 40 value 665.581829
## iter 50 value 665.579022
## iter 50 value 665.579020
## iter 50 value 665.579020
## final value 665.579020
## converged
## # weights: 46
## initial value 1079.438837
## iter 10 value 709.807351
## iter 20 value 697.034476
## iter 30 value 672.335215
## iter 40 value 613.099826
## iter 50 value 480.365172
## iter 60 value 465.124978
## iter 70 value 454.957704
## iter 80 value 454.437292
## final value 454.437285
## converged
## # weights: 67
## initial value 1031.733228
## iter 10 value 668.420997
## iter 20 value 667.639469
## iter 30 value 659.616962
## iter 40 value 611.999336
## iter 50 value 505.634894
## iter 60 value 497.391853
## iter 70 value 490.815418
## iter 80 value 463.127676
## iter 90 value 452.658878
## final value 452.538368
## converged
## # weights: 109
## initial value 1028.916739
## iter 10 value 670.977463
## iter 20 value 646.035332
## iter 30 value 490.566479
## iter 40 value 469.664405
## iter 50 value 444.116587
## iter 60 value 432.960306
## iter 70 value 421.285188
## iter 80 value 413.856101
## iter 90 value 408.979298
## iter 100 value 406.346043
## final value 406.346043
## stopped after 100 iterations
## # weights: 151
## initial value 880.004200
## iter 10 value 670.270854
## iter 20 value 668.744855
## iter 30 value 668.013307
## iter 40 value 656.251359
## iter 50 value 540.946737
## iter 60 value 484.306918
## iter 70 value 470.653201
## iter 80 value 464.333996
## iter 90 value 454.541151
## iter 100 value 452.225293
## final value 452.225293
## stopped after 100 iterations
## # weights: 46
## initial value 964.977497
## iter 10 value 661.133423
## iter 20 value 658.493850
## iter 30 value 658.469934
## iter 40 value 658.357033
## iter 50 value 658.343595
## final value 658.343419
## converged
## # weights: 67
## initial value 1531.308266
## iter 10 value 665.099232
## iter 20 value 663.892976
## iter 30 value 658.027215
## iter 40 value 581.959989
## iter 50 value 511.586323
## iter 60 value 453.642052
## iter 70 value 438.350853
## iter 80 value 418.216801
## iter 90 value 412.280640
## iter 100 value 410.917889
## final value 410.917889
## stopped after 100 iterations
## # weights: 109
## initial value 1263.796691
## iter 10 value 657.944481
## iter 20 value 657.878887
## iter 30 value 657.703826
## iter 40 value 627.861482
## iter 50 value 593.475602
## iter 60 value 510.833699
## iter 70 value 444.160047
## iter 80 value 417.036336
## iter 90 value 412.414392
## iter 100 value 410.587902
## final value 410.587902
## stopped after 100 iterations
## # weights: 151
## initial value 1024.493836
## iter 10 value 667.603355
## iter 20 value 655.438431
## iter 30 value 450.002509
## iter 40 value 429.874600
## iter 50 value 421.560761
## iter 60 value 407.871646
## iter 70 value 373.359127
## iter 80 value 354.123562
## iter 90 value 347.193016
## iter 100 value 346.321688
## final value 346.321688
## stopped after 100 iterations
## # weights: 46
## initial value 1081.310390
## iter 10 value 671.295674
## iter 20 value 661.301281
## iter 30 value 660.734577
## iter 40 value 659.670961
## iter 50 value 659.455845
## final value 659.454591
## converged
## # weights: 67
## initial value 1390.307914
## iter 10 value 665.826474
## iter 20 value 659.254738
## iter 30 value 658.773901
## iter 40 value 658.772251
## final value 658.772238
## converged
## # weights: 109
## initial value 1036.110575
## iter 10 value 686.208639
## iter 20 value 660.414545
## iter 30 value 659.988932
## iter 30 value 659.988929
## iter 40 value 659.904196
## iter 50 value 637.190520
## iter 60 value 522.979497
## iter 70 value 432.146770
## iter 80 value 413.710994
## iter 90 value 393.595965
## iter 100 value 381.346541
## final value 381.346541
## stopped after 100 iterations
## # weights: 151
## initial value 1029.580609
## iter 10 value 661.439377
## iter 20 value 658.795407
## iter 30 value 657.784758
## iter 40 value 657.675580
## iter 50 value 657.667491
## final value 657.667450
## converged
## # weights: 46
## initial value 942.551047
## iter 10 value 670.725087
## iter 20 value 663.084824
## iter 30 value 662.375116
## iter 40 value 643.814142
## iter 50 value 494.879507
## iter 60 value 457.661155
## iter 70 value 441.166777
## iter 80 value 429.146651
## iter 90 value 427.077835
## iter 100 value 426.937268
## final value 426.937268
## stopped after 100 iterations
## # weights: 67
## initial value 755.608832
## iter 10 value 665.442429
## iter 20 value 662.194372
## iter 30 value 662.111778
## iter 40 value 661.577413
## iter 50 value 660.066882
## iter 60 value 659.842260
## iter 70 value 659.707834
## iter 80 value 659.594713
## final value 659.586819
## converged
## # weights: 109
## initial value 1042.759471
## iter 10 value 660.523377
## iter 20 value 659.150862
## iter 30 value 658.793302
## iter 40 value 658.738221
## iter 50 value 658.674761
## iter 60 value 658.320896
## iter 70 value 645.851991
## iter 80 value 465.751673
## iter 90 value 447.258280
## iter 100 value 433.849439
## final value 433.849439
## stopped after 100 iterations
## # weights: 151
## initial value 1116.443093
## iter 10 value 659.043634
## iter 20 value 658.605344
## iter 30 value 658.417304
## iter 40 value 658.345445
## iter 50 value 454.832492
## iter 60 value 429.357577
## iter 70 value 415.860148
## iter 80 value 405.151339
## iter 90 value 390.634884
## iter 100 value 374.107387
## final value 374.107387
## stopped after 100 iterations
## # weights: 151
## initial value 2253.814575
## iter 10 value 1008.855893
## iter 20 value 992.496092
## iter 30 value 990.592991
## iter 40 value 944.778643
## iter 50 value 923.802498
## iter 60 value 904.121902
## iter 70 value 793.454004
## iter 80 value 676.678024
## iter 90 value 667.323414
## iter 100 value 645.657470
## final value 645.657470
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n5_NN1Fit0
## Neural Network
##
## 989 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 659, 661, 658
## Resampling results across tuning parameters:
##
## size decay Accuracy Kappa
## 2 0.3 0.6251961 0.1827544
## 2 0.5 0.6187537 0.1755853
## 2 0.7 0.7240180 0.4723262
## 3 0.3 0.7260290 0.4942634
## 3 0.5 0.6695728 0.3190770
## 3 0.7 0.6211311 0.1679244
## 5 0.3 0.7229956 0.4706652
## 5 0.5 0.7219978 0.4694074
## 5 0.7 0.6684624 0.3086276
## 7 0.3 0.7402044 0.5259517
## 7 0.5 0.6177436 0.1604981
## 7 0.7 0.6684532 0.3100571
##
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were size = 7 and decay = 0.3.
DryBean_TDA_KDE_5.50.5_n5_NN1Fit0$resample
## Accuracy Kappa Resample
## 1 0.7652439 0.5754152 Fold2
## 2 0.7272727 0.4814944 Fold1
## 3 0.7280967 0.5209456 Fold3
nb_tda_kde_5.50.5_n5_nn1_fit_re<-DryBean_TDA_KDE_5.50.5_n5_NN1Fit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n5_NN1Fit0)
## a 16-7-4 network with 151 weights
## options were - softmax modelling decay=0.3
## b->h1 i1->h1 i2->h1 i3->h1 i4->h1 i5->h1 i6->h1 i7->h1 i8->h1 i9->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h1 i11->h1 i12->h1 i13->h1 i14->h1 i15->h1 i16->h1
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h2 i1->h2 i2->h2 i3->h2 i4->h2 i5->h2 i6->h2 i7->h2 i8->h2 i9->h2
## 0.00 0.07 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00
## i10->h2 i11->h2 i12->h2 i13->h2 i14->h2 i15->h2 i16->h2
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h3 i1->h3 i2->h3 i3->h3 i4->h3 i5->h3 i6->h3 i7->h3 i8->h3 i9->h3
## -0.01 -0.02 -0.16 1.13 0.74 -0.02 0.00 0.01 -1.13 0.05
## i10->h3 i11->h3 i12->h3 i13->h3 i14->h3 i15->h3 i16->h3
## -0.01 -0.02 -0.02 0.00 0.00 -0.02 -0.03
## b->h4 i1->h4 i2->h4 i3->h4 i4->h4 i5->h4 i6->h4 i7->h4 i8->h4 i9->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h4 i11->h4 i12->h4 i13->h4 i14->h4 i15->h4 i16->h4
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h5 i1->h5 i2->h5 i3->h5 i4->h5 i5->h5 i6->h5 i7->h5 i8->h5 i9->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h5 i11->h5 i12->h5 i13->h5 i14->h5 i15->h5 i16->h5
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h6 i1->h6 i2->h6 i3->h6 i4->h6 i5->h6 i6->h6 i7->h6 i8->h6 i9->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## i10->h6 i11->h6 i12->h6 i13->h6 i14->h6 i15->h6 i16->h6
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->h7 i1->h7 i2->h7 i3->h7 i4->h7 i5->h7 i6->h7 i7->h7 i8->h7 i9->h7
## 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
## i10->h7 i11->h7 i12->h7 i13->h7 i14->h7 i15->h7 i16->h7
## 0.00 0.00 0.00 0.00 0.00 0.00 0.00
## b->o1 h1->o1 h2->o1 h3->o1 h4->o1 h5->o1 h6->o1 h7->o1
## 0.12 0.12 0.12 2.27 0.17 0.00 0.12 0.11
## b->o2 h1->o2 h2->o2 h3->o2 h4->o2 h5->o2 h6->o2 h7->o2
## -0.62 -0.62 -0.62 0.41 -0.03 0.00 -0.62 -0.58
## b->o3 h1->o3 h2->o3 h3->o3 h4->o3 h5->o3 h6->o3 h7->o3
## 0.56 0.56 0.56 -5.11 -0.19 0.00 0.56 0.53
## b->o4 h1->o4 h2->o4 h3->o4 h4->o4 h5->o4 h6->o4 h7->o4
## -0.07 -0.07 -0.07 2.43 0.05 0.00 -0.07 -0.07
#vip(DryBean_TDA_KDE_5.50.5_n5_NN1Fit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n5_NN1Fit TDA-Assited NN")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n5_NN1Fit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n5_NN1Fit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n5_db_nn1_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n5_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 13 0 91 1038 573 21 662
## HOROZ 0 0 0 0 0 0 0
## SEKER 383 156 398 25 5 587 128
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3983
## 95% CI : (0.3832, 0.4135)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2339
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9765
## Specificity 1.00000 1.00000 1.0000 0.5492
## Pos Pred Value NaN NaN NaN 0.4329
## Neg Pred Value 0.90294 0.96176 0.8801 0.9851
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2544
## Detection Prevalence 0.00000 0.00000 0.0000 0.5877
## Balanced Accuracy 0.50000 0.50000 0.5000 0.7629
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.9655 0.0000
## Specificity 1.0000 0.6846 1.0000
## Pos Pred Value NaN 0.3490 NaN
## Neg Pred Value 0.8583 0.9912 0.8064
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1439 0.0000
## Detection Prevalence 0.0000 0.4123 0.0000
## Balanced Accuracy 0.5000 0.8250 0.5000
nb_tda_kde_5.50.5_n5_db_nn1_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 13 0 91 1038 573 21 662
## HOROZ 0 0 0 0 0 0 0
## SEKER 383 156 398 25 5 587 128
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3983
## 95% CI : (0.3832, 0.4135)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2339
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9765
## Specificity 1.00000 1.00000 1.0000 0.5492
## Pos Pred Value NaN NaN NaN 0.4329
## Neg Pred Value 0.90294 0.96176 0.8801 0.9851
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2544
## Detection Prevalence 0.00000 0.00000 0.0000 0.5877
## Balanced Accuracy 0.50000 0.50000 0.5000 0.7629
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.9655 0.0000
## Specificity 1.0000 0.6846 1.0000
## Pos Pred Value NaN 0.3490 NaN
## Neg Pred Value 0.8583 0.9912 0.8064
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.0000 0.1439 0.0000
## Detection Prevalence 0.0000 0.4123 0.0000
## Balanced Accuracy 0.5000 0.8250 0.5000
nb_tda_kde_5.50.5_n5_db_nn1_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.982843e-01 2.339080e-01 3.832177e-01 4.134967e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 4.962694e-82 NaN
nb_tda_kde_5.50.5_n5_db_nn1_cf0_ov_acc<-nb_tda_kde_5.50.5_n5_db_nn1_cf0$overall[1]
nb_tda_kde_5.50.5_n5_db_nn1_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.9764817 0.5492211 0.4328607 0.9851367 0.4328607
## Class: HOROZ 0.0000000 1.0000000 NaN 0.8583333 NA
## Class: SEKER 0.9654605 0.6846198 0.3489893 0.9912427 0.3489893
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.0000000
## Class: BOMBAY 0.0000000 NA 0.03823529 0.0000000
## Class: CALI 0.0000000 NA 0.11985294 0.0000000
## Class: DERMASON 0.9764817 0.5998266 0.26053922 0.2544118
## Class: HOROZ 0.0000000 NA 0.14166667 0.0000000
## Class: SEKER 0.9654605 0.5126638 0.14901961 0.1438725
## Class: SIRA 0.0000000 NA 0.19362745 0.0000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000 0.5000000
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.0000000 0.5000000
## Class: DERMASON 0.5877451 0.7628514
## Class: HOROZ 0.0000000 0.5000000
## Class: SEKER 0.4122549 0.8250402
## Class: SIRA 0.0000000 0.5000000
nb_tda_kde_5.50.5_n5_db_nn1_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n5_db_nn1_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold<-(db_nn1_fit_re - nb_tda_kde_5.50.5_n5_nn1_fit_re)
diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold
## Accuracy
## 1 -0.05963937
## 2 -0.19406530
## 3 -0.16075873
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n5_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n5_3_fold_odds.left<-bst_tda_kde_5.50.5_nn1.n5_3_fold$probLeft/bst_tda_kde_5.50.5_nn1.n5_3_fold$probRight
bst_tda_kde_5.50.5_nn1.n5_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n5_3_fold
## $winLeft
## [1] 0.9920667
##
## $winRope
## [1] 0.007933333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n5_3_fold
## $left
## [1] 0.9445112
##
## $rope
## [1] 0.01221637
##
## $right
## [1] 0.0432724
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold))
#bf_tda_kde_5.50.5_nn1.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nn1_n5_3_fold)
## t = -3.4182, df = 2, p-value = 0.07596
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.31205825 0.03574932
## sample estimates:
## mean of x
## -0.1381545
### Test set diff
diff_drybean_tda_kde_5.50.5_nn1.n5_test<-(db_nn1_cf_ov_acc - nb_tda_kde_5.50.5_n5_db_nn1_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nn1.n5_test
## Accuracy
## 0.2115196
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n5_test),-0.01,0.01)
bst_tda_kde_5.50.5_nn1.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nn1.n5_test_odds.left<-bst_tda_kde_5.50.5_nn1.n5_test$probLeft/bst_tda_kde_5.50.5_nn1.n5_test$probRight
bst_tda_kde_5.50.5_nn1.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nn1.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nn1.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1615667
##
## $winRight
## [1] 0.8384333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nn1.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n5_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nn1.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nn1.n5_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nn1.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n5_test)) #bf_tda_pca_5.50.5_nn1.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nn1.n5_test))
##Logistic Regression method='multinom'
dryBeanLrFit <- train(as.factor(Class) ~ .,
data = Dry_Bean_DatasetTrain,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 126 (102 variable)
## initial value 12362.367177
## iter 10 value 9330.369345
## iter 20 value 6749.354731
## iter 30 value 5678.621798
## iter 40 value 2359.286993
## iter 50 value 1381.324503
## iter 60 value 1303.958222
## iter 70 value 1280.976922
## iter 80 value 1261.906693
## iter 90 value 1252.052252
## iter 100 value 1243.031033
## final value 1243.031033
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12362.367177
## iter 10 value 9330.369392
## iter 20 value 6749.356862
## iter 30 value 5678.995387
## iter 40 value 2512.636380
## iter 50 value 1725.772824
## iter 60 value 1610.816489
## iter 70 value 1531.708206
## iter 80 value 1482.938511
## iter 90 value 1465.564607
## iter 100 value 1452.820198
## final value 1452.820198
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12362.367177
## iter 10 value 9330.369345
## iter 20 value 6749.354729
## iter 30 value 5678.623940
## iter 40 value 2360.172572
## iter 50 value 1382.779074
## iter 60 value 1307.338376
## iter 70 value 1285.935476
## iter 80 value 1269.615240
## iter 90 value 1262.230024
## iter 100 value 1255.496272
## final value 1255.496272
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12366.258997
## iter 10 value 9324.925181
## iter 20 value 7332.973845
## iter 30 value 5126.001635
## iter 40 value 2638.877179
## iter 50 value 1423.772588
## iter 60 value 1345.259517
## iter 70 value 1328.387799
## iter 80 value 1310.396748
## iter 90 value 1298.268238
## iter 100 value 1291.770919
## final value 1291.770919
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12366.258997
## iter 10 value 9324.925219
## iter 20 value 7332.975459
## iter 30 value 5126.608164
## iter 40 value 2722.252546
## iter 50 value 1777.540654
## iter 60 value 1655.443575
## iter 70 value 1580.415537
## iter 80 value 1528.144994
## iter 90 value 1515.229401
## iter 100 value 1504.569564
## final value 1504.569564
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12366.258997
## iter 10 value 9324.925181
## iter 20 value 7332.973850
## iter 30 value 5126.002046
## iter 40 value 2638.937341
## iter 50 value 1425.421293
## iter 60 value 1348.924858
## iter 70 value 1333.437606
## iter 80 value 1318.416578
## iter 90 value 1309.319649
## iter 100 value 1304.157066
## final value 1304.157066
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12364.313087
## iter 10 value 9303.251690
## iter 20 value 6747.215042
## iter 30 value 5136.956724
## iter 40 value 2710.122161
## iter 50 value 1365.293349
## iter 60 value 1299.474377
## iter 70 value 1281.985811
## iter 80 value 1267.371275
## iter 90 value 1258.602573
## iter 100 value 1245.996684
## final value 1245.996684
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12364.313087
## iter 10 value 9303.251723
## iter 20 value 6747.217164
## iter 30 value 5137.231414
## iter 40 value 2681.655242
## iter 50 value 1698.885165
## iter 60 value 1576.613279
## iter 70 value 1501.036571
## iter 80 value 1440.450343
## iter 90 value 1424.106938
## iter 100 value 1414.444134
## final value 1414.444134
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12364.313087
## iter 10 value 9303.251690
## iter 20 value 6747.215042
## iter 30 value 5136.956830
## iter 40 value 2709.853870
## iter 50 value 1366.718618
## iter 60 value 1302.274050
## iter 70 value 1285.880948
## iter 80 value 1273.350491
## iter 90 value 1266.768135
## iter 100 value 1257.838280
## final value 1257.838280
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 18546.469631
## iter 10 value 14256.077488
## iter 20 value 10801.375725
## iter 30 value 8317.981350
## iter 40 value 4842.299633
## iter 50 value 2094.588199
## iter 60 value 1992.650958
## iter 70 value 1957.386666
## iter 80 value 1940.152053
## iter 90 value 1927.837065
## iter 100 value 1919.271883
## final value 1919.271883
## stopped after 100 iterations
dryBeanLrFit
## Penalized Multinomial Regression
##
## 9531 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6353, 6355, 6354
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9264521 0.9110561
## 1e-04 0.9268719 0.9115603
## 1e-01 0.9224645 0.9062325
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
dryBeanLrFit$resample
## Accuracy Kappa Resample
## 1 0.9363980 0.9230796 Fold2
## 2 0.9203902 0.9037874 Fold1
## 3 0.9238275 0.9078139 Fold3
db_lr_fit_re<-dryBeanLrFit$resample[1]
summary(dryBeanLrFit)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 5.097996 0.001518203 -0.08485485 0.6867606 1.444120
## CALI 27.758816 0.003852931 -0.18871191 2.1712986 2.872933
## DERMASON 22.581674 0.005414543 0.19965082 1.1314684 1.626920
## HOROZ 4.712840 0.008208654 0.09969379 2.5184362 4.258753
## SEKER -8.485302 0.008216020 0.15925215 -0.2937576 -1.380356
## SIRA 66.354878 0.004679263 -0.38854574 2.2510720 2.947044
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 37.28312 7.644902 -9.677267e-06 -2.4349872 -8.321374
## CALI -60.69857 95.595706 -3.981068e-03 -4.2920059 4.056257
## DERMASON -21.95784 74.657652 -3.629310e-03 -4.4236490 -16.330867
## HOROZ -16.65023 97.110516 -7.241419e-03 -7.4567788 -6.170527
## SEKER -22.19845 -48.301470 -7.531226e-03 0.5973303 -11.721895
## SIRA -55.06946 134.255959 -4.635402e-03 -4.0925410 -6.884650
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 7.968895 17.74026 1.854190 0.67565651 0.18278340
## CALI 28.500269 -46.46146 5.614737 0.71126202 0.13176749
## DERMASON 11.057829 139.34613 9.193598 0.27073145 0.01648839
## HOROZ 41.195855 74.66557 -25.228524 0.88101195 -0.21155705
## SEKER -7.785538 97.57366 25.634165 -1.11510060 0.06150360
## SIRA 42.793980 -144.76055 42.479503 0.06103648 -0.02510808
## ShapeFactor3 ShapeFactor4
## BOMBAY 1.028511 8.450051
## CALI -23.183451 -3.648962
## DERMASON -13.521871 5.847190
## HOROZ -58.066290 -4.029509
## SEKER 59.277237 8.638673
## SIRA 3.867839 28.032780
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 7.917461e-06 0.0026703592 0.004074527 0.0015121523 0.0009994635
## CALI 2.831263e-06 0.0003613638 0.001249474 0.0006797879 0.0005053984
## DERMASON 7.794527e-06 0.0007816329 0.002488559 0.0017441752 0.0019341325
## HOROZ 4.011739e-06 0.0004717857 0.001717104 0.0005561916 0.0006173356
## SEKER 4.745019e-06 0.0009470733 0.002047941 0.0005773709 0.0006787073
## SIRA 6.308116e-06 0.0005622078 0.002006143 0.0023566795 0.0024504302
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 1.234439e-05 5.872410e-06 0.0026328242 0.0012264358 5.719707e-06
## CALI 6.003313e-06 2.287693e-06 0.0003583472 0.0003895703 2.388238e-06
## DERMASON 1.956993e-05 7.462169e-06 0.0007898878 0.0008631003 7.606134e-06
## HOROZ 5.134695e-06 2.750474e-06 0.0004665014 0.0005727688 3.068941e-06
## SEKER 5.117149e-06 2.835002e-06 0.0009451085 0.0006541179 3.776227e-06
## SIRA 2.518209e-05 8.250341e-06 0.0005627223 0.0007952691 6.852901e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 7.820727e-06 7.027581e-06 6.393240e-06 6.217948e-08 1.621544e-08
## CALI 2.792485e-06 2.793458e-06 2.751830e-06 2.488082e-08 9.092922e-09
## DERMASON 7.706436e-06 9.218825e-06 1.148968e-05 6.718853e-08 6.240651e-08
## HOROZ 3.945310e-06 3.816346e-06 3.693618e-06 3.169838e-08 1.114481e-08
## SEKER 4.698978e-06 4.455297e-06 4.407947e-06 3.727738e-08 1.494083e-08
## SIRA 6.209524e-06 9.457196e-06 1.267679e-05 3.983808e-08 6.754685e-08
## ShapeFactor3 ShapeFactor4
## BOMBAY 5.192961e-06 7.912661e-06
## CALI 2.787191e-06 2.819754e-06
## DERMASON 1.428358e-05 7.804055e-06
## HOROZ 3.351762e-06 3.985891e-06
## SEKER 4.034923e-06 4.740540e-06
## SIRA 1.667630e-05 6.386398e-06
##
## Residual Deviance: 3838.544
## AIC: 4042.544
vip(dryBeanLrFit,25) + ggtitle('non-TDA-Assisted LR')
# Predict outcome using model from training data based on testing data
predictions <- predict(dryBeanLrFit, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_lr_cf<-confusionMatrix(data=predictions, as.factor(Dry_Bean_DatasetTest$Class))
db_lr_cf
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 17 0 1 8 3
## BOMBAY 0 156 0 0 0 0 0
## CALI 20 0 454 0 11 1 1
## DERMASON 0 0 0 954 6 10 71
## HOROZ 2 0 11 2 552 0 11
## SEKER 2 0 1 21 0 572 2
## SIRA 8 0 6 86 8 17 702
##
## Overall Statistics
##
## Accuracy : 0.9201
## 95% CI : (0.9114, 0.9282)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9034
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9284 0.8975
## Specificity 0.99213 1.00000 0.9908 0.9712
## Pos Pred Value 0.92621 1.00000 0.9322 0.9164
## Neg Pred Value 0.99132 1.00000 0.9903 0.9641
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1113 0.2338
## Detection Prevalence 0.09632 0.03824 0.1194 0.2551
## Balanced Accuracy 0.95566 1.00000 0.9596 0.9343
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9550 0.9408 0.8886
## Specificity 0.9926 0.9925 0.9620
## Pos Pred Value 0.9550 0.9565 0.8489
## Neg Pred Value 0.9926 0.9897 0.9729
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1353 0.1402 0.1721
## Detection Prevalence 0.1417 0.1466 0.2027
## Balanced Accuracy 0.9738 0.9667 0.9253
db_lr_cf$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9200980 0.9034124 0.9113514 0.9282373 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_lr_cf_ov_acc<-db_lr_cf$overall[1]
db_lr_cf$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.9921281 0.9262087 0.9913209 0.9262087
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9284254 0.9908104 0.9322382 0.9902588 0.9322382
## Class: DERMASON 0.8974600 0.9711634 0.9164265 0.9641329 0.9164265
## Class: HOROZ 0.9550173 0.9925757 0.9550173 0.9925757 0.9550173
## Class: SEKER 0.9407895 0.9925115 0.9565217 0.9896611 0.9565217
## Class: SIRA 0.8886076 0.9620061 0.8488513 0.9729480 0.8488513
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.9226869 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9284254 0.9303279 0.11985294 0.11127451
## Class: DERMASON 0.8974600 0.9068441 0.26053922 0.23382353
## Class: HOROZ 0.9550173 0.9550173 0.14166667 0.13529412
## Class: SEKER 0.9407895 0.9485904 0.14901961 0.14019608
## Class: SIRA 0.8886076 0.8682746 0.19362745 0.17205882
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09632353 0.9556600
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.11936275 0.9596179
## Class: DERMASON 0.25514706 0.9343117
## Class: HOROZ 0.14166667 0.9737965
## Class: SEKER 0.14656863 0.9666505
## Class: SIRA 0.20269608 0.9253068
db_lr_cf_pre_rec_f1<-db_lr_cf$byClass[5:7]
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_PC_5.50.5_n1_LrFit0 <- multinom(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n1.vec, family = 'binomial')
## # weights: 108 (85 variable)
## initial value 14045.602479
## iter 10 value 3976.144108
## iter 20 value 3670.317727
## iter 30 value 2455.719076
## iter 40 value 1931.457008
## iter 50 value 1856.281767
## iter 60 value 1829.581353
## iter 70 value 1821.171008
## iter 80 value 1804.889762
## iter 90 value 1799.122621
## iter 100 value 1797.526628
## final value 1797.526628
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n1_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_dry_bean_dataset_5.50.5.n1.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2443.746575
## iter 20 value 2093.043697
## iter 30 value 1587.296150
## iter 40 value 1290.380255
## iter 50 value 1245.870952
## iter 60 value 1222.553446
## iter 70 value 1217.809803
## iter 80 value 1201.856647
## iter 90 value 1195.943338
## iter 100 value 1193.414757
## final value 1193.414757
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2443.748044
## iter 20 value 2093.055139
## iter 30 value 1623.256265
## iter 40 value 1332.519927
## iter 50 value 1319.578901
## iter 60 value 1319.150857
## iter 70 value 1318.996536
## final value 1318.996118
## converged
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2443.746576
## iter 20 value 2093.043714
## iter 30 value 1587.304683
## iter 40 value 1290.640806
## iter 50 value 1247.181782
## iter 60 value 1225.794821
## iter 70 value 1221.655723
## iter 80 value 1209.663244
## iter 90 value 1206.200492
## iter 100 value 1204.878911
## final value 1204.878911
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2566.846410
## iter 20 value 2260.870551
## iter 30 value 1641.776203
## iter 40 value 1292.790518
## iter 50 value 1230.502926
## iter 60 value 1200.099350
## iter 70 value 1192.915489
## iter 80 value 1182.844068
## iter 90 value 1174.130184
## iter 100 value 1171.460590
## final value 1171.460590
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2566.847825
## iter 20 value 2260.874011
## iter 30 value 1783.076382
## iter 40 value 1331.819176
## iter 50 value 1309.140681
## iter 60 value 1308.059692
## iter 70 value 1307.792999
## final value 1307.785534
## converged
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2566.846411
## iter 20 value 2260.870562
## iter 30 value 1642.689583
## iter 40 value 1295.918669
## iter 50 value 1236.654500
## iter 60 value 1205.186961
## iter 70 value 1197.730223
## iter 80 value 1189.876552
## iter 90 value 1184.099164
## iter 100 value 1182.278289
## final value 1182.278289
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2458.954630
## iter 20 value 2072.499864
## iter 30 value 1409.906716
## iter 40 value 1277.434269
## iter 50 value 1232.353128
## iter 60 value 1219.141190
## iter 70 value 1215.590912
## iter 80 value 1209.665427
## iter 90 value 1204.289742
## iter 100 value 1201.576611
## final value 1201.576611
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2458.956084
## iter 20 value 2072.511111
## iter 30 value 1435.173320
## iter 40 value 1333.168473
## iter 50 value 1329.555932
## iter 60 value 1328.989434
## iter 70 value 1328.898097
## final value 1328.898001
## converged
## # weights: 108 (85 variable)
## initial value 9363.734986
## iter 10 value 2458.954632
## iter 20 value 2072.499886
## iter 30 value 1409.952236
## iter 40 value 1277.841337
## iter 50 value 1234.546414
## iter 60 value 1222.545718
## iter 70 value 1219.467621
## iter 80 value 1215.109436
## iter 90 value 1211.586692
## iter 100 value 1210.299762
## final value 1210.299762
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 14045.602479
## iter 10 value 3976.144108
## iter 20 value 3670.317727
## iter 30 value 2455.719076
## iter 40 value 1931.457008
## iter 50 value 1856.281767
## iter 60 value 1829.581353
## iter 70 value 1821.171008
## iter 80 value 1804.889762
## iter 90 value 1799.122621
## iter 100 value 1797.526628
## final value 1797.526628
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n1_LrFit0
## Penalized Multinomial Regression
##
## 7839 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5226, 5226, 5226
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9100651 0.8614578
## 1e-04 0.9095548 0.8606863
## 1e-01 0.9057278 0.8546027
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 0.
DryBean_TDA_PC_5.50.5_n1_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9054726 0.8540440 Fold3
## 2 0.9100651 0.8616784 Fold2
## 3 0.9146575 0.8686511 Fold1
db_tda_pc_5.50.5_n1_lr_fit_re<-DryBean_TDA_PC_5.50.5_n1_LrFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI -6.805172 0.014524813 -0.002344032 -3.6305275 -5.106682
## DERMASON 21.142337 0.003319246 0.214427739 1.4746151 1.826362
## HOROZ -11.413209 -0.001031752 -0.174726641 1.4669020 2.989412
## SEKER -33.635268 0.007845274 0.296875870 -0.2400706 -2.167316
## SIRA 54.976640 0.002473839 -0.365733129 2.7140917 3.183017
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI 0.2363331 -16.05648 -0.016098460 9.296873 -43.48817
## DERMASON -14.4931013 27.19918 -0.002846655 -4.524010 -34.52531
## HOROZ 17.7647076 49.15071 -0.003389686 -2.387465 -31.79814
## SEKER -39.1032462 -97.33689 -0.007449825 0.928775 -30.27408
## SIRA -70.2706885 115.31302 -0.002561091 -4.814693 -25.62583
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI -6.713114 -18.33349 -4.193017 -0.11033846 -0.01564704
## DERMASON -19.211617 141.63227 11.189794 0.74861814 0.44729411
## HOROZ -9.718945 -29.79388 -39.588675 0.07322316 -0.47407662
## SEKER -18.943963 150.17355 30.755231 -1.47096929 0.13169631
## SIRA 71.746834 -139.74119 40.366393 0.77262522 0.28817003
## ShapeFactor3 ShapeFactor4
## CALI -0.3147478 -4.208654
## DERMASON -5.1900489 -3.468595
## HOROZ -63.3798463 -24.910382
## SEKER 93.6830863 8.790040
## SIRA 10.2798160 32.025067
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI 6.242985e-07 0.0001022191 0.000297086 0.0001044604 5.629651e-05
## DERMASON 6.696232e-06 0.0009880656 0.001919046 0.0020467490 2.104018e-03
## HOROZ 6.179209e-06 0.0020296692 0.002577041 0.0012009467 3.048743e-04
## SEKER 4.570700e-06 0.0008911461 0.001977499 0.0008096503 4.195887e-04
## SIRA 5.854455e-06 0.0009315765 0.001913747 0.0021066623 2.007447e-03
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI 9.939293e-07 4.998684e-07 0.0001000448 7.738052e-05 4.389490e-07
## DERMASON 2.306598e-05 8.607583e-06 0.0009840845 7.250028e-04 6.745788e-06
## HOROZ 1.457193e-05 6.069319e-06 0.0020115505 6.609513e-04 4.647886e-06
## SEKER 7.488201e-06 4.002717e-06 0.0008882299 5.648664e-04 3.422807e-06
## SIRA 2.305160e-05 8.665284e-06 0.0009284368 6.707533e-04 6.100129e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI 6.163512e-07 4.893486e-07 4.910909e-07 5.954112e-09 1.548476e-09
## DERMASON 6.617547e-06 8.796041e-06 1.180564e-05 6.230080e-08 6.752508e-08
## HOROZ 6.115574e-06 4.607653e-06 3.819712e-06 8.155191e-08 9.524541e-09
## SEKER 4.522457e-06 4.041766e-06 3.660598e-06 4.370569e-08 1.329759e-08
## SIRA 5.786258e-06 7.784422e-06 1.084717e-05 5.465057e-08 6.181417e-08
## ShapeFactor3 ShapeFactor4
## CALI 3.842223e-07 6.232035e-07
## DERMASON 1.550949e-05 6.720790e-06
## HOROZ 2.227296e-06 6.175856e-06
## SEKER 3.041702e-06 4.557710e-06
## SIRA 1.449632e-05 5.881349e-06
##
## Residual Deviance: 3595.053
## AIC: 3765.053
vip(DryBean_TDA_PC_5.50.5_n1_LrFit0,50) + ggtitle("dryBean_TDA_PCA_5.50.5_n1_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_PC_5.50.5_n1_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n1_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n1_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n1_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 150 29 12 0 0 5 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 963 31 10 75
## HOROZ 0 0 0 1 21 0 0
## SEKER 224 127 165 21 6 575 3
## SIRA 22 0 312 78 520 18 712
##
## Overall Statistics
##
## Accuracy : 0.5934
## 95% CI : (0.5781, 0.6085)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4954
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.37879 0.00000 0.0000 0.9059
## Specificity 0.98751 1.00000 1.0000 0.9616
## Pos Pred Value 0.76531 NaN NaN 0.8925
## Neg Pred Value 0.93666 0.96176 0.8801 0.9667
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.03676 0.00000 0.0000 0.2360
## Detection Prevalence 0.04804 0.00000 0.0000 0.2645
## Balanced Accuracy 0.68315 0.50000 0.5000 0.9337
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.036332 0.9457 0.9013
## Specificity 0.999714 0.8427 0.7112
## Pos Pred Value 0.954545 0.5129 0.4284
## Neg Pred Value 0.862740 0.9888 0.9677
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.005147 0.1409 0.1745
## Detection Prevalence 0.005392 0.2748 0.4074
## Balanced Accuracy 0.518023 0.8942 0.8063
db_tda_pc_5.50.5_n1_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 150 29 12 0 0 5 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 963 31 10 75
## HOROZ 0 0 0 1 21 0 0
## SEKER 224 127 165 21 6 575 3
## SIRA 22 0 312 78 520 18 712
##
## Overall Statistics
##
## Accuracy : 0.5934
## 95% CI : (0.5781, 0.6085)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.4954
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.37879 0.00000 0.0000 0.9059
## Specificity 0.98751 1.00000 1.0000 0.9616
## Pos Pred Value 0.76531 NaN NaN 0.8925
## Neg Pred Value 0.93666 0.96176 0.8801 0.9667
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.03676 0.00000 0.0000 0.2360
## Detection Prevalence 0.04804 0.00000 0.0000 0.2645
## Balanced Accuracy 0.68315 0.50000 0.5000 0.9337
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.036332 0.9457 0.9013
## Specificity 0.999714 0.8427 0.7112
## Pos Pred Value 0.954545 0.5129 0.4284
## Neg Pred Value 0.862740 0.9888 0.9677
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.005147 0.1409 0.1745
## Detection Prevalence 0.005392 0.2748 0.4074
## Balanced Accuracy 0.518023 0.8942 0.8063
db_tda_pc_5.50.5_n1_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.5933824 0.4954193 0.5781251 0.6085058 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n1_db_lr_cf0_ov_acc<-db_tda_pc_5.50.5_n1_db_lr_cf0$overall[1]
db_tda_pc_5.50.5_n1_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.37878788 0.9875136 0.7653061 0.9366632 0.7653061
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.00000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.90592662 0.9615512 0.8924930 0.9666778 0.8924930
## Class: HOROZ 0.03633218 0.9997144 0.9545455 0.8627403 0.9545455
## Class: SEKER 0.94572368 0.8427419 0.5129349 0.9888476 0.5129349
## Class: SIRA 0.90126582 0.7112462 0.4283995 0.9677419 0.4283995
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.37878788 0.5067568 0.09705882 0.036764706
## Class: BOMBAY 0.00000000 NA 0.03823529 0.000000000
## Class: CALI 0.00000000 NA 0.11985294 0.000000000
## Class: DERMASON 0.90592662 0.8991597 0.26053922 0.236029412
## Class: HOROZ 0.03633218 0.0700000 0.14166667 0.005147059
## Class: SEKER 0.94572368 0.6651243 0.14901961 0.140931373
## Class: SIRA 0.90126582 0.5807504 0.19362745 0.174509804
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.048039216 0.6831507
## Class: BOMBAY 0.000000000 0.5000000
## Class: CALI 0.000000000 0.5000000
## Class: DERMASON 0.264460784 0.9337389
## Class: HOROZ 0.005392157 0.5180233
## Class: SEKER 0.274754902 0.8942328
## Class: SIRA 0.407352941 0.8062560
db_tda_pc_5.50.5_n1_db_lr_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n1_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted LR vs. tda-assisted LR classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_lr_n1_3_fold<-(db_lr_fit_re - db_tda_pc_5.50.5_n1_lr_fit_re)
diff_drybean_tda_pca_5.50.5_lr_n1_3_fold
## Accuracy
## 1 0.030925348
## 2 0.010325123
## 3 0.009170028
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n1_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n1_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.3933667
##
## $winRight
## [1] 0.6066333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n1_3_fold
## $left
## [1] 0.04075021
##
## $rope
## [1] 0.2052336
##
## $right
## [1] 0.7540162
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold))
#bf_tda_pca_5.50.5_lr.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_lr_n1_3_fold)
## t = 2.3782, df = 2, p-value = 0.1405
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.01360057 0.04721423
## sample estimates:
## mean of x
## 0.01680683
### Test set diff
diff_drybean_tda_pca_5.50.5_lr.n1_test<-(db_lr_cf_ov_acc - db_tda_pc_5.50.5_n1_db_lr_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_lr.n1_test
## Accuracy
## 0.3267157
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n1_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n1_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n1_test$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n1_test$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n1_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1614333
##
## $winRight
## [1] 0.8385667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n1_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr.n1_test)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr.n1_test)) #bf_tda_pca_5.50.5_lr.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n1_test))
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node2
DryBean_TDA_PC_5.50.5_n2_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 126 (102 variable)
## initial value 12340.962165
## iter 10 value 7074.107699
## iter 20 value 5048.312161
## iter 30 value 4504.574761
## iter 40 value 2156.660814
## iter 50 value 1801.015774
## iter 60 value 1754.533217
## iter 70 value 1713.391756
## iter 80 value 1701.170874
## iter 90 value 1693.732301
## iter 100 value 1690.816821
## final value 1690.816821
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12340.962165
## iter 10 value 7074.107947
## iter 20 value 5048.314715
## iter 30 value 4504.631013
## iter 40 value 2337.731267
## iter 50 value 2000.012357
## iter 60 value 1899.856826
## iter 70 value 1861.319722
## iter 80 value 1854.397204
## iter 90 value 1853.657119
## iter 100 value 1853.137230
## final value 1853.137230
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12340.962165
## iter 10 value 7074.107699
## iter 20 value 5048.312170
## iter 30 value 4504.575836
## iter 40 value 2156.118730
## iter 50 value 1802.238357
## iter 60 value 1757.702008
## iter 70 value 1721.197308
## iter 80 value 1711.207258
## iter 90 value 1705.165078
## iter 100 value 1702.772523
## final value 1702.772523
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12344.853986
## iter 10 value 7001.946358
## iter 20 value 4750.975536
## iter 30 value 4281.897814
## iter 40 value 2163.324259
## iter 50 value 1714.916709
## iter 60 value 1669.282838
## iter 70 value 1628.082414
## iter 80 value 1608.317145
## iter 90 value 1602.889075
## iter 100 value 1598.384910
## final value 1598.384910
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12344.853986
## iter 10 value 7001.946604
## iter 20 value 4750.978616
## iter 30 value 4281.965880
## iter 40 value 2334.883371
## iter 50 value 1922.990025
## iter 60 value 1796.780394
## iter 70 value 1770.984678
## iter 80 value 1757.537897
## iter 90 value 1756.410932
## iter 100 value 1756.080671
## final value 1756.080671
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 12344.853986
## iter 10 value 7001.946358
## iter 20 value 4750.975541
## iter 30 value 4281.898263
## iter 40 value 2163.550168
## iter 50 value 1716.120744
## iter 60 value 1672.127829
## iter 70 value 1634.777585
## iter 80 value 1618.411878
## iter 90 value 1614.270289
## iter 100 value 1610.811768
## final value 1610.811768
## stopped after 100 iterations
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'BOMBAY' is empty
## # weights: 108 (85 variable)
## initial value 11366.922073
## iter 10 value 6113.426403
## iter 20 value 4232.554444
## iter 30 value 2990.316145
## iter 40 value 1871.381646
## iter 50 value 1779.900414
## iter 60 value 1738.983473
## iter 70 value 1721.231790
## iter 80 value 1708.645138
## iter 90 value 1703.861873
## iter 100 value 1700.556767
## final value 1700.556767
## stopped after 100 iterations
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'BOMBAY' is empty
## # weights: 108 (85 variable)
## initial value 11366.922073
## iter 10 value 6113.426819
## iter 20 value 4232.565129
## iter 30 value 3043.738911
## iter 40 value 2064.143828
## iter 50 value 1930.220649
## iter 60 value 1874.194879
## iter 70 value 1856.809189
## iter 80 value 1853.912580
## iter 90 value 1853.482291
## iter 100 value 1852.860992
## final value 1852.860992
## stopped after 100 iterations
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'BOMBAY' is empty
## # weights: 108 (85 variable)
## initial value 11366.922073
## iter 10 value 6113.426404
## iter 20 value 4232.554444
## iter 30 value 2990.361566
## iter 40 value 1872.102435
## iter 50 value 1782.303111
## iter 60 value 1743.910679
## iter 70 value 1727.936735
## iter 80 value 1717.297448
## iter 90 value 1713.504586
## iter 100 value 1710.909472
## final value 1710.909472
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 18515.335068
## iter 10 value 11489.490912
## iter 20 value 9045.153506
## iter 30 value 7988.113055
## iter 40 value 4033.414254
## iter 50 value 2687.884270
## iter 60 value 2603.938106
## iter 70 value 2545.459568
## iter 80 value 2528.875693
## iter 90 value 2520.071258
## iter 100 value 2514.556975
## final value 2514.556975
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n2_LrFit0
## Penalized Multinomial Regression
##
## 9515 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6342, 6344, 6344
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9004721 0.8743309
## 1e-04 0.9003668 0.8741878
## 1e-01 0.8962682 0.8690413
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 0.
DryBean_TDA_PC_5.50.5_n2_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.8990855 0.8725637 Fold3
## 2 0.8978240 0.8710404 Fold2
## 3 0.9045068 0.8793886 Fold1
db_tda_pc_5.50.5_n2_lr_fit_re<-DryBean_TDA_PC_5.50.5_n2_LrFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n2_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 16.074628 0.001721945 -0.09625761 0.3037480 0.6718628
## CALI 30.918418 0.003068693 -0.20276057 2.2234406 2.7252793
## DERMASON -7.635020 0.004152185 0.24362292 0.8490522 1.5633786
## HOROZ 9.005719 0.007163367 0.06060321 2.5190182 4.0217133
## SEKER -29.580169 0.004796938 0.20455498 0.2908672 -1.0448602
## SIRA 65.101267 0.002586321 -0.44240350 2.3919875 2.6001153
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 28.885094 19.18656 0.0006097655 -1.6291048 1.645970
## CALI -81.578458 102.80003 -0.0037830854 -3.9311565 2.183730
## DERMASON -9.005589 42.84656 -0.0042641550 -3.3872747 -15.042047
## HOROZ -27.582461 85.88235 -0.0066387273 -6.9230333 -5.605022
## SEKER -64.516377 -47.51655 -0.0066522959 0.4765184 -11.675516
## SIRA -88.449657 110.36087 -0.0047891150 -2.9035403 -6.990231
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 15.82905 16.82593 10.317682 0.2955878 0.04388263
## CALI 30.52384 -53.04375 3.815354 0.8162554 0.07676899
## DERMASON -13.97817 152.66681 -17.920757 -0.4401114 -0.39111534
## HOROZ 40.50633 55.29302 -20.973157 0.1465449 -0.27879455
## SEKER -42.38644 117.39610 16.912407 -0.8957035 0.33345150
## SIRA 38.61104 -175.32440 39.984941 1.2832901 0.34394405
## ShapeFactor3 ShapeFactor4
## BOMBAY 5.567188 15.143927
## CALI -28.636315 -6.393010
## DERMASON -32.207763 -15.426964
## HOROZ -53.305330 -4.177148
## SEKER 56.829461 -12.958122
## SIRA -1.007753 19.674859
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 8.166560e-08 2.653592e-05 3.413936e-05 1.349225e-05 7.683018e-06
## CALI 2.702519e-06 3.717195e-04 1.255802e-03 6.229623e-04 4.586995e-04
## DERMASON 7.238461e-06 6.915100e-04 2.407938e-03 1.474467e-03 1.742129e-03
## HOROZ 3.922213e-06 5.571448e-04 1.697389e-03 6.335806e-04 5.280845e-04
## SEKER 5.258025e-06 7.280802e-04 2.237899e-03 6.496242e-04 6.509797e-04
## SIRA 5.617260e-06 5.303947e-04 1.983924e-03 2.118285e-03 2.116032e-03
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 1.392699e-07 6.559283e-08 2.680686e-05 1.016571e-05 6.150724e-08
## CALI 5.278181e-06 2.115903e-06 3.704401e-04 3.851650e-04 2.111512e-06
## DERMASON 1.643336e-05 6.028305e-06 7.048440e-04 8.054496e-04 7.118182e-06
## HOROZ 5.822946e-06 2.897865e-06 5.582560e-04 5.561792e-04 2.722683e-06
## SEKER 6.158551e-06 3.250045e-06 7.342073e-04 6.617309e-04 4.192633e-06
## SIRA 2.241037e-05 7.429005e-06 5.366989e-04 6.909851e-04 6.302452e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 8.057899e-08 7.115880e-08 6.267958e-08 8.370317e-10 1.825037e-10
## CALI 2.673500e-06 2.596532e-06 2.591972e-06 2.203705e-08 8.408772e-09
## DERMASON 7.154692e-06 8.078973e-06 1.040691e-05 5.935106e-08 5.428715e-08
## HOROZ 3.872327e-06 3.723584e-06 3.368814e-06 3.256429e-08 9.730858e-09
## SEKER 5.202535e-06 4.703929e-06 4.776175e-06 4.517527e-08 1.687011e-08
## SIRA 5.523577e-06 8.127544e-06 1.109684e-05 3.653772e-08 5.982842e-08
## ShapeFactor3 ShapeFactor4
## BOMBAY 4.821937e-08 8.146744e-08
## CALI 2.576877e-06 2.713459e-06
## DERMASON 1.266472e-05 7.259569e-06
## HOROZ 2.910692e-06 3.907919e-06
## SEKER 4.296500e-06 5.252031e-06
## SIRA 1.458906e-05 5.708986e-06
##
## Residual Deviance: 5029.114
## AIC: 5233.114
vip(DryBean_TDA_PC_5.50.5_n2_LrFit0,50) + ggtitle("dryBean_TDA_PCA_5.50.5_n2_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_PC_5.50.5_n2_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n2_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n2_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n2_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 16 1 1 7 3
## BOMBAY 1 156 0 0 0 0 0
## CALI 19 0 454 0 9 0 3
## DERMASON 0 0 0 963 5 12 69
## HOROZ 3 0 11 3 553 0 8
## SEKER 2 0 1 16 0 572 4
## SIRA 7 0 7 80 10 17 703
##
## Overall Statistics
##
## Accuracy : 0.9228
## 95% CI : (0.9142, 0.9308)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9067
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9284 0.9059
## Specificity 0.99240 0.99975 0.9914 0.9715
## Pos Pred Value 0.92857 0.99363 0.9361 0.9180
## Neg Pred Value 0.99132 1.00000 0.9903 0.9670
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1113 0.2360
## Detection Prevalence 0.09608 0.03848 0.1189 0.2571
## Balanced Accuracy 0.95580 0.99987 0.9599 0.9387
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9408 0.8899
## Specificity 0.9929 0.9934 0.9632
## Pos Pred Value 0.9567 0.9613 0.8532
## Neg Pred Value 0.9929 0.9897 0.9733
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1402 0.1723
## Detection Prevalence 0.1417 0.1458 0.2020
## Balanced Accuracy 0.9748 0.9671 0.9265
db_tda_pc_5.50.5_n2_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 16 1 1 7 3
## BOMBAY 1 156 0 0 0 0 0
## CALI 19 0 454 0 9 0 3
## DERMASON 0 0 0 963 5 12 69
## HOROZ 3 0 11 3 553 0 8
## SEKER 2 0 1 16 0 572 4
## SIRA 7 0 7 80 10 17 703
##
## Overall Statistics
##
## Accuracy : 0.9228
## 95% CI : (0.9142, 0.9308)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9067
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9284 0.9059
## Specificity 0.99240 0.99975 0.9914 0.9715
## Pos Pred Value 0.92857 0.99363 0.9361 0.9180
## Neg Pred Value 0.99132 1.00000 0.9903 0.9670
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1113 0.2360
## Detection Prevalence 0.09608 0.03848 0.1189 0.2571
## Balanced Accuracy 0.95580 0.99987 0.9599 0.9387
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9408 0.8899
## Specificity 0.9929 0.9934 0.9632
## Pos Pred Value 0.9567 0.9613 0.8532
## Neg Pred Value 0.9929 0.9897 0.9733
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1402 0.1723
## Detection Prevalence 0.1417 0.1458 0.2020
## Balanced Accuracy 0.9748 0.9671 0.9265
db_tda_pc_5.50.5_n2_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9227941 0.9066505 0.9141749 0.9308019 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n2_db_lr_cf0_ov_acc<-db_tda_pc_5.50.5_n2_db_lr_cf0$overall[1]
db_tda_pc_5.50.5_n2_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.9923996 0.9285714 0.9913232 0.9285714
## Class: BOMBAY 1.0000000 0.9997452 0.9936306 1.0000000 0.9936306
## Class: CALI 0.9284254 0.9913673 0.9360825 0.9902643 0.9360825
## Class: DERMASON 0.9059266 0.9714949 0.9180172 0.9670076 0.9180172
## Class: HOROZ 0.9567474 0.9928612 0.9567474 0.9928612 0.9567474
## Class: SEKER 0.9407895 0.9933756 0.9613445 0.9896700 0.9613445
## Class: SIRA 0.8898734 0.9632219 0.8531553 0.9732801 0.8531553
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.9238579 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 0.9968051 0.03823529 0.03823529
## Class: CALI 0.9284254 0.9322382 0.11985294 0.11127451
## Class: DERMASON 0.9059266 0.9119318 0.26053922 0.23602941
## Class: HOROZ 0.9567474 0.9567474 0.14166667 0.13553922
## Class: SEKER 0.9407895 0.9509559 0.14901961 0.14019608
## Class: SIRA 0.8898734 0.8711276 0.19362745 0.17230392
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09607843 0.9557957
## Class: BOMBAY 0.03848039 0.9998726
## Class: CALI 0.11887255 0.9598963
## Class: DERMASON 0.25710784 0.9387107
## Class: HOROZ 0.14166667 0.9748043
## Class: SEKER 0.14583333 0.9670825
## Class: SIRA 0.20196078 0.9265477
db_tda_pc_5.50.5_n2_db_lr_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n2_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted LR vs. tda-assisted LR classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_lr_n2_3_fold<-(db_lr_fit_re - db_tda_pc_5.50.5_n2_lr_fit_re)
diff_drybean_tda_pca_5.50.5_lr_n2_3_fold
## Accuracy
## 1 0.03731252
## 2 0.02256615
## 3 0.01932073
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n2_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n2_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.03696667
##
## $winRight
## [1] 0.9630333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n2_3_fold
## $left
## [1] 0.01474312
##
## $rope
## [1] 0.04738248
##
## $right
## [1] 0.9378744
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold))
#bf_tda_pca_5.50.5_lr.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_lr_n2_3_fold)
## t = 4.7686, df = 2, p-value = 0.04127
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.002579426 0.050220181
## sample estimates:
## mean of x
## 0.0263998
### Test set diff
diff_drybean_tda_pca_5.50.5_lr.n2_test<-(db_lr_cf_ov_acc - db_tda_pc_5.50.5_n2_db_lr_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_lr.n2_test
## Accuracy
## -0.002696078
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n2_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 1
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n2_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n2_test$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n2_test$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n2_test_odds.left
## [1] NaN
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n2_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 1
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n2_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr.n2_test)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr.n2_test)) #bf_tda_pca_5.50.5_lr.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n2_test))
##Node3
DryBean_TDA_PC_5.50.5_n3_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4514.997621
## iter 20 value 2951.751507
## iter 30 value 2436.112764
## iter 40 value 801.001707
## iter 50 value 556.327157
## iter 60 value 541.977532
## iter 70 value 534.175569
## iter 80 value 527.790947
## iter 90 value 524.316187
## iter 100 value 522.121591
## final value 522.121591
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4514.997698
## iter 20 value 2951.765476
## iter 30 value 2436.367014
## iter 40 value 873.878393
## iter 50 value 647.130147
## iter 60 value 605.776183
## iter 70 value 590.751820
## iter 80 value 588.105743
## iter 90 value 586.443548
## iter 100 value 585.006086
## final value 585.006086
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4514.997624
## iter 20 value 2951.752914
## iter 30 value 2436.147309
## iter 40 value 800.901379
## iter 50 value 556.791295
## iter 60 value 542.832332
## iter 70 value 535.818817
## iter 80 value 531.017896
## iter 90 value 528.668068
## iter 100 value 527.366451
## final value 527.366451
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 3869.823349
## iter 20 value 2201.134171
## iter 30 value 1887.819960
## iter 40 value 812.503341
## iter 50 value 587.866312
## iter 60 value 569.347098
## iter 70 value 558.741299
## iter 80 value 550.564403
## iter 90 value 542.504291
## iter 100 value 539.417432
## final value 539.417432
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 3869.823410
## iter 20 value 2201.145829
## iter 30 value 1887.955038
## iter 40 value 964.306843
## iter 50 value 653.220706
## iter 60 value 619.668800
## iter 70 value 611.611444
## iter 80 value 609.407498
## iter 90 value 608.222845
## iter 100 value 607.270427
## final value 607.270427
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 3869.823349
## iter 20 value 2201.134175
## iter 30 value 1887.820453
## iter 40 value 812.681840
## iter 50 value 588.460774
## iter 60 value 570.490550
## iter 70 value 560.891150
## iter 80 value 554.014623
## iter 90 value 547.908858
## iter 100 value 545.984264
## final value 545.984264
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4236.257154
## iter 20 value 2607.394196
## iter 30 value 2130.988976
## iter 40 value 816.099481
## iter 50 value 564.068075
## iter 60 value 551.110729
## iter 70 value 542.163309
## iter 80 value 537.312876
## iter 90 value 530.721770
## iter 100 value 525.261849
## final value 525.261849
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4236.257165
## iter 20 value 2607.399227
## iter 30 value 2131.144300
## iter 40 value 897.212419
## iter 50 value 654.262126
## iter 60 value 613.218418
## iter 70 value 592.419456
## iter 80 value 589.809006
## iter 90 value 589.026648
## iter 100 value 587.602478
## final value 587.602478
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 6946.899232
## iter 10 value 4236.257154
## iter 20 value 2607.394249
## iter 30 value 2130.996984
## iter 40 value 816.975911
## iter 50 value 564.904001
## iter 60 value 552.174775
## iter 70 value 544.096230
## iter 80 value 540.082746
## iter 90 value 535.062639
## iter 100 value 532.237558
## final value 532.237558
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10420.348848
## iter 10 value 6845.310431
## iter 20 value 3500.938717
## iter 30 value 2783.052684
## iter 40 value 1069.786186
## iter 50 value 874.975492
## iter 60 value 844.382610
## iter 70 value 828.321111
## iter 80 value 815.153614
## iter 90 value 807.463694
## iter 100 value 802.936198
## final value 802.936198
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n3_LrFit0
## Penalized Multinomial Regression
##
## 5355 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 3570, 3570, 3570
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9449113 0.9232484
## 1e-04 0.9447246 0.9229828
## 1e-01 0.9415500 0.9184811
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 0.
DryBean_TDA_PC_5.50.5_n3_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9450980 0.9237105 Fold3
## 2 0.9439776 0.9218613 Fold2
## 3 0.9456583 0.9241733 Fold1
db_tda_pc_5.50.5_n3_lr_fit_re<-DryBean_TDA_PC_5.50.5_n2_LrFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n3_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 9.85067 0.009119522 -0.09002700 1.169633 2.586884
## CALI 21.37667 0.002922220 -0.17027119 1.753240 2.632045
## DERMASON 22.21659 0.003894321 -0.01368438 1.089894 1.943034
## HOROZ -13.79197 0.006834102 0.10253300 1.745168 4.070059
## SEKER 34.04545 0.002625139 -0.05243008 1.237072 1.783588
## SIRA 14.84737 0.001692132 0.00573875 1.986698 1.726498
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 56.69264 18.78195 -0.007113250 -4.135310 -2.918904
## CALI -58.76140 130.59714 -0.003342126 -3.494826 3.857366
## DERMASON 35.33526 14.83545 -0.001742719 -3.979289 5.784793
## HOROZ 30.97531 88.83514 -0.006420320 -6.030582 -5.149266
## SEKER -6.48801 33.72721 -0.001193855 -3.537147 -35.851309
## SIRA -97.22407 58.72235 -0.004063682 -3.187320 -9.216045
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 6.934488 -1.169424 -0.01176916 0.2343526 -0.03927711
## CALI 31.091798 -40.278082 -17.89230207 0.4244595 -0.27333353
## DERMASON 17.589892 1.429623 19.06426277 0.4377699 0.12210271
## HOROZ 8.536009 79.062409 -57.10920381 0.3085262 -0.51018483
## SEKER 19.186947 12.556114 29.66440801 0.4275961 0.17717002
## SIRA 12.789659 71.467933 -0.59003747 -0.1299115 -0.13237123
## ShapeFactor3 ShapeFactor4
## BOMBAY -5.373915 9.751554
## CALI -60.381695 -15.088440
## DERMASON 16.658093 23.049217
## HOROZ -97.473939 -28.792974
## SEKER 20.573700 21.470160
## SIRA -23.906360 -14.138014
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 8.670993e-07 0.0001788435 0.0004640906 1.849404e-04 9.020876e-05
## CALI 9.401235e-06 0.0002998468 0.0010982449 3.424343e-03 4.093801e-03
## DERMASON 1.095476e-07 0.0021392502 0.0002046423 4.544822e-05 5.100414e-05
## HOROZ 4.527463e-06 0.0003823974 0.0019055272 2.083366e-03 1.838130e-03
## SEKER 5.022275e-07 0.0009646133 0.0002550029 9.421915e-05 4.384243e-05
## SIRA 6.105328e-06 0.0005528918 0.0028354980 9.592361e-04 6.566459e-04
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 1.731974e-06 7.340871e-07 0.0001701765 1.282997e-04 6.511723e-07
## CALI 3.085479e-05 4.581312e-06 0.0002978887 1.385375e-03 8.791476e-06
## DERMASON 5.720006e-07 1.467320e-07 0.0020332314 6.326258e-06 7.036579e-08
## HOROZ 1.949012e-05 4.010235e-06 0.0003794985 6.622067e-04 3.308819e-06
## SEKER 9.591364e-07 4.388559e-07 0.0009436872 6.149589e-05 3.219585e-07
## SIRA 1.004507e-05 4.829997e-06 0.0005591507 7.738328e-04 4.397460e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 8.532193e-07 6.876297e-07 6.187827e-07 8.004086e-09 1.316262e-09
## CALI 9.263770e-06 1.529659e-05 1.730632e-05 1.772266e-08 6.477991e-08
## DERMASON 1.516780e-08 2.379575e-07 2.349288e-07 1.147431e-10 1.317496e-09
## HOROZ 4.462579e-06 6.233295e-06 7.719062e-06 3.167198e-08 2.918906e-08
## SEKER 4.757019e-07 3.606177e-07 3.616405e-07 5.296474e-09 1.013266e-09
## SIRA 6.016073e-06 4.838463e-06 4.891159e-06 5.933683e-08 1.388876e-08
## ShapeFactor3 ShapeFactor4
## BOMBAY 4.454696e-07 8.648184e-07
## CALI 2.100637e-05 9.204588e-06
## DERMASON 2.990672e-07 3.627933e-08
## HOROZ 9.419990e-06 4.435812e-06
## SEKER 2.630352e-07 4.871964e-07
## SIRA 3.969653e-06 6.147876e-06
##
## Residual Deviance: 1605.872
## AIC: 1809.872
vip(DryBean_TDA_PC_5.50.5_n3_LrFit0,50) + ggtitle("dryBean_TDA_PCA_5.50.5_n3_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_PC_5.50.5_n3_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n3_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n3_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
db_tda_pc_5.50.5_n3_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 366 0 17 1 1 22 5
## BOMBAY 0 154 0 0 0 0 0
## CALI 17 0 458 0 10 0 3
## DERMASON 0 2 0 66 0 0 0
## HOROZ 4 0 10 2 553 0 11
## SEKER 1 0 0 189 0 359 0
## SIRA 8 0 4 805 14 227 771
##
## Overall Statistics
##
## Accuracy : 0.6684
## 95% CI : (0.6537, 0.6828)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6067
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.92424 0.98718 0.9366 0.06209
## Specificity 0.98751 1.00000 0.9916 0.99934
## Pos Pred Value 0.88835 1.00000 0.9385 0.97059
## Neg Pred Value 0.99182 0.99949 0.9914 0.75150
## Prevalence 0.09706 0.03824 0.1199 0.26054
## Detection Rate 0.08971 0.03775 0.1123 0.01618
## Detection Prevalence 0.10098 0.03775 0.1196 0.01667
## Balanced Accuracy 0.95588 0.99359 0.9641 0.53071
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.59046 0.9759
## Specificity 0.9923 0.94528 0.6784
## Pos Pred Value 0.9534 0.65392 0.4215
## Neg Pred Value 0.9929 0.92948 0.9916
## Prevalence 0.1417 0.14902 0.1936
## Detection Rate 0.1355 0.08799 0.1890
## Detection Prevalence 0.1422 0.13456 0.4483
## Balanced Accuracy 0.9745 0.76787 0.8272
db_tda_pc_5.50.5_n3_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 366 0 17 1 1 22 5
## BOMBAY 0 154 0 0 0 0 0
## CALI 17 0 458 0 10 0 3
## DERMASON 0 2 0 66 0 0 0
## HOROZ 4 0 10 2 553 0 11
## SEKER 1 0 0 189 0 359 0
## SIRA 8 0 4 805 14 227 771
##
## Overall Statistics
##
## Accuracy : 0.6684
## 95% CI : (0.6537, 0.6828)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6067
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.92424 0.98718 0.9366 0.06209
## Specificity 0.98751 1.00000 0.9916 0.99934
## Pos Pred Value 0.88835 1.00000 0.9385 0.97059
## Neg Pred Value 0.99182 0.99949 0.9914 0.75150
## Prevalence 0.09706 0.03824 0.1199 0.26054
## Detection Rate 0.08971 0.03775 0.1123 0.01618
## Detection Prevalence 0.10098 0.03775 0.1196 0.01667
## Balanced Accuracy 0.95588 0.99359 0.9641 0.53071
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.59046 0.9759
## Specificity 0.9923 0.94528 0.6784
## Pos Pred Value 0.9534 0.65392 0.4215
## Neg Pred Value 0.9929 0.92948 0.9916
## Prevalence 0.1417 0.14902 0.1936
## Detection Rate 0.1355 0.08799 0.1890
## Detection Prevalence 0.1422 0.13456 0.4483
## Balanced Accuracy 0.9745 0.76787 0.8272
db_tda_pc_5.50.5_n3_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6683824 0.6066627 0.6536971 0.6828261 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n3_db_lr_cf0_ov_acc<-db_tda_pc_5.50.5_n3_db_lr_cf0$overall[1]
db_tda_pc_5.50.5_n3_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.92424242 0.9875136 0.8883495 0.9918212 0.8883495
## Class: BOMBAY 0.98717949 1.0000000 1.0000000 0.9994906 1.0000000
## Class: CALI 0.93660532 0.9916458 0.9385246 0.9913697 0.9385246
## Class: DERMASON 0.06208843 0.9993371 0.9705882 0.7514955 0.9705882
## Class: HOROZ 0.95674740 0.9922901 0.9534483 0.9928571 0.9534483
## Class: SEKER 0.59046053 0.9452765 0.6539162 0.9294817 0.6539162
## Class: SIRA 0.97594937 0.6784195 0.4215418 0.9915593 0.4215418
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.92424242 0.9059406 0.09705882 0.08970588
## Class: BOMBAY 0.98717949 0.9935484 0.03823529 0.03774510
## Class: CALI 0.93660532 0.9375640 0.11985294 0.11225490
## Class: DERMASON 0.06208843 0.1167109 0.26053922 0.01617647
## Class: HOROZ 0.95674740 0.9550950 0.14166667 0.13553922
## Class: SEKER 0.59046053 0.6205704 0.14901961 0.08799020
## Class: SIRA 0.97594937 0.5887743 0.19362745 0.18897059
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.10098039 0.9558780
## Class: BOMBAY 0.03774510 0.9935897
## Class: CALI 0.11960784 0.9641255
## Class: DERMASON 0.01666667 0.5307128
## Class: HOROZ 0.14215686 0.9745188
## Class: SEKER 0.13455882 0.7678685
## Class: SIRA 0.44828431 0.8271844
db_tda_pc_5.50.5_n3_db_lr_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n3_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted LR vs. tda-assisted LR classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_lr_n3_3_fold<-(db_lr_fit_re - db_tda_pc_5.50.5_n3_lr_fit_re)
diff_drybean_tda_pca_5.50.5_lr_n3_3_fold
## Accuracy
## 1 0.03731252
## 2 0.02256615
## 3 0.01932073
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n3_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n3_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.0377
##
## $winRight
## [1] 0.9623
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n3_3_fold
## $left
## [1] 0.01474312
##
## $rope
## [1] 0.04738248
##
## $right
## [1] 0.9378744
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold))
#bf_tda_pca_5.50.5_lr.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_lr_n3_3_fold)
## t = 4.7686, df = 2, p-value = 0.04127
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.002579426 0.050220181
## sample estimates:
## mean of x
## 0.0263998
### Test set diff
diff_drybean_tda_pca_5.50.5_lr.n3_test<-(db_lr_cf_ov_acc - db_tda_pc_5.50.5_n3_db_lr_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_lr.n3_test
## Accuracy
## 0.2517157
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n3_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n3_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n3_test$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n3_test$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n3_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n3_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n3_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1568667
##
## $winRight
## [1] 0.8431333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n3_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr.n3_test)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr.n3_test)) #bf_tda_pca_5.50.5_lr.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n3_test))
##Node4
DryBean_TDA_PC_5.50.5_n4_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 72 (51 variable)
## initial value 1470.858317
## iter 10 value 618.079432
## iter 20 value 212.550525
## iter 30 value 50.363671
## iter 40 value 48.230025
## iter 50 value 47.343623
## iter 60 value 46.789835
## iter 70 value 46.542332
## iter 80 value 46.361652
## iter 90 value 46.259890
## iter 100 value 45.920749
## final value 45.920749
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1470.858317
## iter 10 value 618.079493
## iter 20 value 215.975148
## iter 30 value 80.350523
## iter 40 value 62.019205
## iter 50 value 57.853443
## iter 60 value 57.514581
## iter 70 value 57.498470
## iter 80 value 57.491022
## iter 90 value 57.476253
## iter 100 value 57.423924
## final value 57.423924
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1470.858317
## iter 10 value 618.079432
## iter 20 value 212.552510
## iter 30 value 50.599813
## iter 40 value 48.477229
## iter 50 value 47.578097
## iter 60 value 47.064273
## iter 70 value 46.886634
## iter 80 value 46.801554
## iter 90 value 46.740556
## iter 100 value 46.501255
## final value 46.501255
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1468.085728
## iter 10 value 650.446414
## iter 20 value 140.642786
## iter 30 value 55.600797
## iter 40 value 52.581913
## iter 50 value 51.503603
## iter 60 value 50.824243
## iter 70 value 50.004078
## iter 80 value 48.626175
## iter 90 value 47.929465
## iter 100 value 47.830153
## final value 47.830153
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1468.085728
## iter 10 value 650.446464
## iter 20 value 178.001562
## iter 30 value 83.794914
## iter 40 value 65.150422
## iter 50 value 61.157328
## iter 60 value 59.987595
## iter 70 value 59.713364
## iter 80 value 59.677665
## iter 90 value 59.667228
## iter 100 value 59.661004
## final value 59.661004
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1468.085728
## iter 10 value 650.446414
## iter 20 value 140.681139
## iter 30 value 55.790161
## iter 40 value 52.836606
## iter 50 value 51.825845
## iter 60 value 51.234492
## iter 70 value 50.630508
## iter 80 value 49.966557
## iter 90 value 49.744692
## iter 100 value 49.715354
## final value 49.715354
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1469.472023
## iter 10 value 658.560214
## iter 20 value 182.787195
## iter 30 value 68.150202
## iter 40 value 65.944292
## iter 50 value 64.388886
## iter 60 value 63.609920
## iter 70 value 62.805373
## iter 80 value 62.175524
## iter 90 value 61.879703
## iter 100 value 61.045030
## final value 61.045030
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1469.472023
## iter 10 value 658.560402
## iter 20 value 186.360584
## iter 30 value 93.889368
## iter 40 value 78.692496
## iter 50 value 72.713399
## iter 60 value 72.319048
## iter 70 value 72.243787
## iter 80 value 72.239189
## iter 90 value 72.237709
## iter 100 value 72.230356
## final value 72.230356
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1469.472023
## iter 10 value 658.560214
## iter 20 value 182.793716
## iter 30 value 68.305306
## iter 40 value 66.104387
## iter 50 value 64.620994
## iter 60 value 63.968998
## iter 70 value 63.475439
## iter 80 value 63.134330
## iter 90 value 62.947955
## iter 100 value 62.654666
## final value 62.654666
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 2204.208034
## iter 10 value 783.542866
## iter 20 value 258.362081
## iter 30 value 89.910550
## iter 40 value 86.689336
## iter 50 value 84.599279
## iter 60 value 83.543152
## iter 70 value 83.047437
## iter 80 value 82.871730
## iter 90 value 82.350165
## iter 100 value 82.223466
## final value 82.223466
## stopped after 100 iterations
DryBean_TDA_PC_5.50.5_n4_LrFit0
## Penalized Multinomial Regression
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1061, 1059, 1060
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9805043 0.9715922
## 1e-04 0.9836478 0.9761486
## 1e-01 0.9805043 0.9716044
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_PC_5.50.5_n4_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9811676 0.9724551 Fold2
## 2 0.9810964 0.9723895 Fold1
## 3 0.9886792 0.9836012 Fold3
db_tda_pc_5.50.5_n4_lr_fit_re<-DryBean_TDA_PC_5.50.5_n4_LrFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n4_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY -4.159853 0.0010347292 -0.16514095 0.8722229 1.756878
## CALI -0.793122 0.0006979468 -0.09189237 1.0563761 1.124534
## HOROZ 12.082874 0.0069799208 0.03994281 2.1578960 3.246367
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 23.82217 -3.695187 -0.0001276862 -2.285295 18.547615
## CALI -46.94882 40.309977 -0.0011919297 -1.663021 7.800909
## HOROZ 10.09107 -24.254603 -0.0062027749 -5.921399 -8.331375
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2 ShapeFactor3
## BOMBAY -3.853775 -6.341031 -5.713175 0.08007980 -0.02300731 -4.593631
## CALI 21.718682 17.432279 -14.886856 -0.17804373 -0.14973099 -31.699405
## HOROZ 1.016797 24.252263 26.552978 0.07184372 0.19499819 37.344327
## ShapeFactor4
## BOMBAY -2.145152
## CALI -6.046346
## HOROZ 19.001400
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 7.018313e-07 0.0052243255 0.0008108686 1.744398e-05 0.0002134523
## CALI 4.674780e-06 0.0006076085 0.0031843724 1.422169e-03 0.0003467775
## HOROZ 6.289482e-06 0.0007130044 0.0052699597 1.474966e-03 0.0004951484
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 1.490428e-07 1.228230e-07 0.0051306014 0.0001281071 2.643921e-07
## CALI 1.158052e-05 4.592335e-06 0.0006205885 0.0007541547 3.049354e-06
## HOROZ 1.417523e-05 5.851883e-06 0.0007373128 0.0009089062 5.275184e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 6.036271e-07 1.313798e-07 8.152171e-07 3.192879e-09 2.151808e-09
## CALI 4.586799e-06 3.038004e-06 2.799510e-06 4.224410e-08 4.363133e-09
## HOROZ 6.123367e-06 2.324847e-06 4.143589e-06 6.082113e-08 7.965245e-09
## ShapeFactor3 ShapeFactor4
## BOMBAY 8.671634e-07 6.635749e-07
## CALI 1.586586e-06 4.608469e-06
## HOROZ 2.684442e-06 6.420998e-06
##
## Residual Deviance: 164.4469
## AIC: 266.4469
vip(DryBean_TDA_PC_5.50.5_n4_LrFit0,50) + ggtitle("dryBean_TDA_PCA_5.50.5_n4_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_PC_5.50.5_n4_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n4_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n4_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n4_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 314 0 5 0 1 1 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 32 0 472 0 14 1 28
## DERMASON 0 0 0 0 0 0 0
## HOROZ 50 0 12 1063 563 606 762
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3689
## 95% CI : (0.354, 0.3839)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2735
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.79293 1.00000 0.9652 0.0000
## Specificity 0.99810 1.00000 0.9791 1.0000
## Pos Pred Value 0.97819 1.00000 0.8629 NaN
## Neg Pred Value 0.97819 1.00000 0.9952 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.07696 0.03824 0.1157 0.0000
## Detection Prevalence 0.07868 0.03824 0.1341 0.0000
## Balanced Accuracy 0.89551 1.00000 0.9722 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9740 0.000 0.0000
## Specificity 0.2881 1.000 1.0000
## Pos Pred Value 0.1842 NaN NaN
## Neg Pred Value 0.9854 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1380 0.000 0.0000
## Detection Prevalence 0.7490 0.000 0.0000
## Balanced Accuracy 0.6311 0.500 0.5000
db_tda_pc_5.50.5_n4_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 314 0 5 0 1 1 0
## BOMBAY 0 156 0 0 0 0 0
## CALI 32 0 472 0 14 1 28
## DERMASON 0 0 0 0 0 0 0
## HOROZ 50 0 12 1063 563 606 762
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3689
## 95% CI : (0.354, 0.3839)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2735
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.79293 1.00000 0.9652 0.0000
## Specificity 0.99810 1.00000 0.9791 1.0000
## Pos Pred Value 0.97819 1.00000 0.8629 NaN
## Neg Pred Value 0.97819 1.00000 0.9952 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.07696 0.03824 0.1157 0.0000
## Detection Prevalence 0.07868 0.03824 0.1341 0.0000
## Balanced Accuracy 0.89551 1.00000 0.9722 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9740 0.000 0.0000
## Specificity 0.2881 1.000 1.0000
## Pos Pred Value 0.1842 NaN NaN
## Neg Pred Value 0.9854 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1380 0.000 0.0000
## Detection Prevalence 0.7490 0.000 0.0000
## Balanced Accuracy 0.6311 0.500 0.5000
db_tda_pc_5.50.5_n4_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.688725e-01 2.734991e-01 3.540431e-01 3.838900e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 2.551451e-52 NaN
db_tda_pc_5.50.5_n4_db_lr_cf0_ov_acc<-db_tda_pc_5.50.5_n4_db_lr_cf0$overall[1]
db_tda_pc_5.50.5_n4_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.7929293 0.9980999 0.9781931 0.9781857 0.9781931
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9652352 0.9791145 0.8628885 0.9951882 0.8628885
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9740484 0.2881211 0.1842277 0.9853516 0.1842277
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.7929293 0.8758717 0.09705882 0.07696078
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9652352 0.9111969 0.11985294 0.11568627
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9740484 0.3098514 0.14166667 0.13799020
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.07867647 0.8955146
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.13406863 0.9721748
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.74901961 0.6310848
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
db_tda_pc_5.50.5_n4_db_lr_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n4_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted LR vs. tda-assisted LR classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_lr_n4_3_fold<-(db_lr_fit_re - db_tda_pc_5.50.5_n4_lr_fit_re)
diff_drybean_tda_pca_5.50.5_lr_n4_3_fold
## Accuracy
## 1 -0.04476962
## 2 -0.06070623
## 3 -0.06485174
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n4_3_fold
## $winLeft
## [1] 0.9918667
##
## $winRope
## [1] 0.008133333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n4_3_fold
## $left
## [1] 0.9889599
##
## $rope
## [1] 0.005530419
##
## $right
## [1] 0.00550964
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold))
#bf_tda_pca_5.50.5_lr.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_lr_n4_3_fold)
## t = -9.2752, df = 2, p-value = 0.01143
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.08311342 -0.03043830
## sample estimates:
## mean of x
## -0.05677586
### Test set diff
diff_drybean_tda_pca_5.50.5_lr.n4_test<-(db_lr_cf_ov_acc - db_tda_pc_5.50.5_n4_db_lr_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_lr.n4_test
## Accuracy
## 0.5512255
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n4_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n4_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n4_test$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n4_test$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n4_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1607667
##
## $winRight
## [1] 0.8392333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n4_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr.n4_test)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr.n4_test)) #bf_tda_pca_5.50.5_lr.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n4_test))
##Node5
DryBean_TDA_PC_5.50.5_n5_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 18 (17 variable)
## initial value 192.694916
## final value 0.000000
## converged
## # weights: 18 (17 variable)
## initial value 192.694916
## iter 10 value 0.001391
## iter 20 value 0.001282
## iter 30 value 0.001280
## iter 40 value 0.000407
## iter 50 value 0.000321
## final value 0.000321
## converged
## # weights: 18 (17 variable)
## initial value 192.694916
## final value 0.000037
## converged
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'CALI' is empty
## Warning: model fit failed for Fold2: decay=0e+00 Error in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## need two or more classes to fit a multinom model
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'CALI' is empty
## Warning: model fit failed for Fold2: decay=1e-01 Error in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## need two or more classes to fit a multinom model
## Warning in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## group 'CALI' is empty
## Warning: model fit failed for Fold2: decay=1e-04 Error in nnet::multinom(.outcome ~ ., data = dat, decay = param$decay, :
## need two or more classes to fit a multinom model
## # weights: 18 (17 variable)
## initial value 192.694916
## final value 184.206807
## converged
## # weights: 18 (17 variable)
## initial value 192.694916
## final value 184.206808
## converged
## # weights: 18 (17 variable)
## initial value 192.694916
## final value 184.206807
## converged
## Warning in nominalTrainWorkflow(x = x, y = y, wts = weights, info = trainInfo,
## : There were missing values in resampled performance measures.
## # weights: 18 (17 variable)
## initial value 289.042374
## final value 184.206807
## converged
DryBean_TDA_PC_5.50.5_n5_LrFit0
## Penalized Multinomial Regression
##
## 417 samples
## 16 predictor
## 2 classes: 'BOMBAY', 'CALI'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 278, 278, 278
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 1.0000000 NaN
## 1e-04 1.0000000 NaN
## 1e-01 0.9928058 0
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_PC_5.50.5_n5_LrFit0$resample
## Accuracy Kappa Resample
## 1 NA NA Fold2
## 2 1 NA Fold1
## 3 1 NA Fold3
db_tda_pc_5.50.5_n5_lr_fit_re<-DryBean_TDA_PC_5.50.5_n5_LrFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n5_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## Values Std. Err.
## (Intercept) -3.399680e-08 0
## Area -6.143051e-03 0
## Perimeter -5.523928e-05 0
## MajorAxisLength -2.079644e-05 0
## MinorAxisLength -1.287831e-05 0
## AspectRation -5.497611e-08 0
## Eccentricity -2.659165e-08 0
## ConvexArea -6.229317e-03 0
## EquivDiameter -1.628301e-05 0
## Extent -2.639320e-08 0
## Solidity -3.352785e-08 0
## roundness -2.917612e-08 0
## Compactness -2.665608e-08 0
## ShapeFactor1 -1.156720e-10 0
## ShapeFactor2 -2.706018e-11 0
## ShapeFactor3 -2.092300e-08 0
## ShapeFactor4 -3.369234e-08 0
##
## Residual Deviance: 368.4136
## AIC: 402.4136
vip(DryBean_TDA_PC_5.50.5_n5_LrFit0,50) + ggtitle("dryBean_TDA_PCA_5.50.5_n5_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_PC_5.50.5_n5_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n5_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n5_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n5_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 396 156 489 1063 578 608 790
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 0 0 0 0
## HOROZ 0 0 0 0 0 0 0
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.0382
## 95% CI : (0.0326, 0.0446)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : 1
##
## Kappa : 0
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 1.00000 0.0000 0.0000
## Specificity 1.00000 0.00000 1.0000 1.0000
## Pos Pred Value NaN 0.03824 NaN NaN
## Neg Pred Value 0.90294 NaN 0.8801 0.7395
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.03824 0.0000 0.0000
## Detection Prevalence 0.00000 1.00000 0.0000 0.0000
## Balanced Accuracy 0.50000 0.50000 0.5000 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.0000 0.000 0.0000
## Specificity 1.0000 1.000 1.0000
## Pos Pred Value NaN NaN NaN
## Neg Pred Value 0.8583 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.0000 0.000 0.0000
## Detection Prevalence 0.0000 0.000 0.0000
## Balanced Accuracy 0.5000 0.500 0.5000
db_tda_pc_5.50.5_n5_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.03823529 0.00000000 0.03256139 0.04458199 0.26053922
## AccuracyPValue McnemarPValue
## 1.00000000 NaN
db_tda_pc_5.50.5_n5_db_lr_cf0_ov_acc<-db_tda_pc_5.50.5_n5_db_lr_cf0$overall[1]
db_tda_pc_5.50.5_n5_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0 1 NaN 0.9029412
## Class: BOMBAY 1 0 0.03823529 NaN
## Class: CALI 0 1 NaN 0.8801471
## Class: DERMASON 0 1 NaN 0.7394608
## Class: HOROZ 0 1 NaN 0.8583333
## Class: SEKER 0 1 NaN 0.8509804
## Class: SIRA 0 1 NaN 0.8063725
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA NA 0 NA 0.09705882 0.00000000
## Class: BOMBAY 0.03823529 1 0.07365439 0.03823529 0.03823529
## Class: CALI NA 0 NA 0.11985294 0.00000000
## Class: DERMASON NA 0 NA 0.26053922 0.00000000
## Class: HOROZ NA 0 NA 0.14166667 0.00000000
## Class: SEKER NA 0 NA 0.14901961 0.00000000
## Class: SIRA NA 0 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0 0.5
## Class: BOMBAY 1 0.5
## Class: CALI 0 0.5
## Class: DERMASON 0 0.5
## Class: HOROZ 0 0.5
## Class: SEKER 0 0.5
## Class: SIRA 0 0.5
db_tda_pc_5.50.5_n5_db_lr_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n5_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted LR vs. tda-assisted LR classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_lr_n5_3_fold<-(db_lr_fit_re - db_tda_pc_5.50.5_n5_lr_fit_re)
diff_drybean_tda_pca_5.50.5_lr_n5_3_fold
## Accuracy
## 1 NA
## 2 -0.07960982
## 3 -0.07617249
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold
## $probLeft
## [1] NA
##
## $probRope
## [1] NA
##
## $probRight
## [1] NA
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n5_3_fold_odds.left
## [1] NA
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_lr.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_lr.n5_3_fold
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n5_3_fold
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold))
#bf_tda_pca_5.50.5_lr.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_lr_n5_3_fold)
## t = -45.321, df = 1, p-value = 0.01404
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.09972885 -0.05605346
## sample estimates:
## mean of x
## -0.07789115
### Test set diff
diff_drybean_tda_pca_5.50.5_lr.n5_test<-(db_lr_cf_ov_acc - db_tda_pc_5.50.5_n5_db_lr_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_lr.n5_test
## Accuracy
## 0.8818627
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n5_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_lr.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_lr.n5_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_lr.n5_test$probLeft/bst_dbf_db_tda_pca_5.50.5_lr.n5_test$probRight
bst_dbf_db_tda_pca_5.50.5_lr.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_lr.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n5_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_lr.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1602
##
## $winRight
## [1] 0.8398
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_lr.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n5_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_lr.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_lr.n5_test)))
#BayesFactor
#bf_tda_pca_5.50.5_lr.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_lr.n5_test)) #bf_tda_pca_5.50.5_lr.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_lr.n5_test))
##With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_KDE_5.50.5_n1_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_kde_dry_bean_dataset_5.50.5.n1.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 126 (102 variable)
## initial value 10988.554612
## iter 10 value 6837.690948
## iter 20 value 5197.858707
## iter 30 value 4209.312091
## iter 40 value 1588.101774
## iter 50 value 900.856802
## iter 60 value 841.534537
## iter 70 value 818.666032
## iter 80 value 807.255799
## iter 90 value 795.197426
## iter 100 value 786.705103
## final value 786.705103
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10988.554612
## iter 10 value 6837.691015
## iter 20 value 5197.860532
## iter 30 value 4209.563304
## iter 40 value 1702.766466
## iter 50 value 1108.844291
## iter 60 value 1039.176591
## iter 70 value 959.172242
## iter 80 value 922.085946
## iter 90 value 909.933562
## iter 100 value 906.120155
## final value 906.120155
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10988.554612
## iter 10 value 6837.690948
## iter 20 value 5197.858711
## iter 30 value 4209.312450
## iter 40 value 1586.117791
## iter 50 value 901.819532
## iter 60 value 843.439990
## iter 70 value 822.156196
## iter 80 value 812.558876
## iter 90 value 803.375843
## iter 100 value 797.650438
## final value 797.650438
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10990.500522
## iter 10 value 6893.066216
## iter 20 value 4959.773884
## iter 30 value 3973.003781
## iter 40 value 2414.116122
## iter 50 value 854.241676
## iter 60 value 803.884228
## iter 70 value 788.864232
## iter 80 value 775.103786
## iter 90 value 768.758086
## iter 100 value 762.963093
## final value 762.963093
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10990.500522
## iter 10 value 6893.066273
## iter 20 value 4959.775140
## iter 30 value 3973.123995
## iter 40 value 2500.463966
## iter 50 value 1109.926310
## iter 60 value 1058.351838
## iter 70 value 1014.756570
## iter 80 value 974.159128
## iter 90 value 961.379358
## iter 100 value 941.668954
## final value 941.668954
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10990.500522
## iter 10 value 6893.066216
## iter 20 value 4959.773878
## iter 30 value 3973.002791
## iter 40 value 2414.278588
## iter 50 value 854.954671
## iter 60 value 805.558223
## iter 70 value 791.551980
## iter 80 value 779.816334
## iter 90 value 775.106310
## iter 100 value 770.972937
## final value 770.972937
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10996.338252
## iter 10 value 6734.360911
## iter 20 value 4820.347477
## iter 30 value 4160.737233
## iter 40 value 1881.503611
## iter 50 value 836.139438
## iter 60 value 776.249227
## iter 70 value 758.117590
## iter 80 value 744.524377
## iter 90 value 737.976497
## iter 100 value 731.936459
## final value 731.936459
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10996.338252
## iter 10 value 6734.361028
## iter 20 value 4820.349545
## iter 30 value 4160.780152
## iter 40 value 1930.494559
## iter 50 value 1064.628679
## iter 60 value 978.157816
## iter 70 value 906.251174
## iter 80 value 862.279139
## iter 90 value 853.042726
## iter 100 value 848.223491
## final value 848.223491
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 10996.338252
## iter 10 value 6734.360911
## iter 20 value 4820.347486
## iter 30 value 4160.737667
## iter 40 value 1881.554777
## iter 50 value 852.516910
## iter 60 value 781.564651
## iter 70 value 763.695852
## iter 80 value 751.532314
## iter 90 value 745.735728
## iter 100 value 741.850397
## final value 741.850397
## stopped after 100 iterations
## # weights: 126 (102 variable)
## initial value 16487.696693
## iter 10 value 12073.546982
## iter 20 value 8017.939505
## iter 30 value 6790.979763
## iter 40 value 2670.846437
## iter 50 value 1325.405930
## iter 60 value 1232.715991
## iter 70 value 1210.574996
## iter 80 value 1186.961362
## iter 90 value 1177.774537
## iter 100 value 1172.339193
## final value 1172.339193
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n1_LrFit0
## Penalized Multinomial Regression
##
## 8473 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5647, 5648, 5651
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9529095 0.9437453
## 1e-04 0.9532635 0.9441686
## 1e-01 0.9498405 0.9400844
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_KDE_5.50.5_n1_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9518584 0.9424883 Fold2
## 2 0.9539986 0.9450324 Fold1
## 3 0.9539334 0.9449850 Fold3
nb_tda_kde_5.50.5_n1_lr_fit_re<-DryBean_TDA_KDE_5.50.5_n1_LrFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n1_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 11.036690 0.003539723 -0.12632891 0.4795659 1.1733521
## CALI 34.788467 0.003118151 -0.18248782 1.9035078 2.5206426
## DERMASON 43.062469 0.007187764 0.07702549 1.3461987 1.7237278
## HOROZ 2.790146 0.007612553 0.08490787 2.1186404 4.0792536
## SEKER -2.394236 0.004046423 0.17706303 0.6937010 0.1230065
## SIRA 42.315250 0.004875151 -0.13696761 2.1870636 2.3826037
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 31.261544 19.68799 -0.001477253 -2.001664 -0.6262265
## CALI -67.559062 108.02139 -0.003336913 -3.652179 4.2446686
## DERMASON -36.404278 92.96568 -0.004909342 -4.561369 -9.9686716
## HOROZ 2.212735 100.14038 -0.006732933 -6.688530 -3.0552608
## SEKER -49.458266 11.38793 -0.003534883 -1.888427 -5.9326146
## SIRA -85.909361 121.14643 -0.005221789 -4.280277 -5.8219881
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 11.37351 9.612811 4.258541 0.2795171 0.003449938
## CALI 39.29064 -50.104213 -1.844882 1.0517992 0.068985682
## DERMASON 39.71568 73.623679 27.991769 0.6917159 0.169481021
## HOROZ 28.47648 61.729422 -34.797416 0.9648725 -0.361563728
## SEKER -22.21867 117.497351 25.948910 -0.8604716 0.069155764
## SIRA 35.06845 -29.357783 31.166003 -0.8037414 -0.154315025
## ShapeFactor3 ShapeFactor4
## BOMBAY -1.002353 10.783763
## CALI -40.888530 -6.805262
## DERMASON 1.393097 21.077324
## HOROZ -73.224152 -9.965704
## SEKER 47.274995 2.244941
## SIRA 1.318080 15.599805
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## BOMBAY 9.318041e-09 0.0002152853 1.846957e-05 2.495589e-05 0.0000130445
## CALI 7.659164e-06 0.0002909245 1.078672e-03 2.538639e-03 0.0031595386
## DERMASON 1.096565e-05 0.0012649178 3.746415e-03 1.443026e-03 0.0014920154
## HOROZ 3.134598e-06 0.0003709058 1.834321e-03 7.222759e-04 0.0006482841
## SEKER 4.817368e-06 0.0007388128 2.477149e-03 6.015784e-04 0.0006498713
## SIRA 6.819379e-06 0.0004611934 2.160322e-03 2.686613e-03 0.0026962499
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## BOMBAY 1.572731e-07 6.282629e-08 0.0002124159 2.092990e-06 1.922534e-08
## CALI 2.323675e-05 3.697050e-06 0.0002885588 1.122317e-03 7.379594e-06
## DERMASON 1.661914e-05 6.857915e-06 0.0012915125 1.202551e-03 9.028714e-06
## HOROZ 7.984153e-06 2.692018e-06 0.0003697209 4.803448e-04 2.688085e-06
## SEKER 5.315782e-06 3.022554e-06 0.0007476451 6.445913e-04 3.759532e-06
## SIRA 2.700959e-05 7.505602e-06 0.0004665447 8.683308e-04 7.475066e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## BOMBAY 1.277472e-08 3.966882e-08 3.069586e-08 2.490169e-10 1.395315e-10
## CALI 7.550245e-06 1.237620e-05 1.369620e-05 1.827909e-08 5.216299e-08
## DERMASON 1.086189e-05 1.075051e-05 1.107203e-05 1.084990e-07 5.167712e-08
## HOROZ 3.071049e-06 3.061709e-06 2.914602e-06 3.411292e-08 7.489913e-09
## SEKER 4.757292e-06 4.203744e-06 4.510602e-06 4.034340e-08 1.871658e-08
## SIRA 6.707306e-06 1.038645e-05 1.307908e-05 4.420472e-08 6.234767e-08
## ShapeFactor3 ShapeFactor4
## BOMBAY 5.580891e-08 9.172119e-09
## CALI 1.658362e-05 7.548106e-06
## DERMASON 1.138473e-05 1.094934e-05
## HOROZ 2.891440e-06 3.127780e-06
## SEKER 4.207713e-06 4.816465e-06
## SIRA 1.680679e-05 6.876640e-06
##
## Residual Deviance: 2344.678
## AIC: 2548.678
vip(DryBean_TDA_KDE_5.50.5_n1_LrFit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n1_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n1_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n1_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n1_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
nb_tda_kde_5.50.5_n1_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 15 0 0 6 2
## BOMBAY 0 156 0 0 0 0 0
## CALI 18 0 458 0 10 0 2
## DERMASON 0 0 0 893 6 7 41
## HOROZ 3 0 8 1 553 0 11
## SEKER 3 0 1 23 0 576 6
## SIRA 8 0 7 146 9 19 728
##
## Overall Statistics
##
## Accuracy : 0.9137
## 95% CI : (0.9047, 0.9222)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8959
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9366 0.8401
## Specificity 0.99376 1.00000 0.9916 0.9821
## Pos Pred Value 0.94057 1.00000 0.9385 0.9430
## Neg Pred Value 0.99133 1.00000 0.9914 0.9457
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1123 0.2189
## Detection Prevalence 0.09485 0.03824 0.1196 0.2321
## Balanced Accuracy 0.95647 1.00000 0.9641 0.9111
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9474 0.9215
## Specificity 0.9934 0.9905 0.9426
## Pos Pred Value 0.9601 0.9458 0.7939
## Neg Pred Value 0.9929 0.9908 0.9804
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1412 0.1784
## Detection Prevalence 0.1412 0.1493 0.2248
## Balanced Accuracy 0.9751 0.9689 0.9320
nb_tda_kde_5.50.5_n1_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 15 0 0 6 2
## BOMBAY 0 156 0 0 0 0 0
## CALI 18 0 458 0 10 0 2
## DERMASON 0 0 0 893 6 7 41
## HOROZ 3 0 8 1 553 0 11
## SEKER 3 0 1 23 0 576 6
## SIRA 8 0 7 146 9 19 728
##
## Overall Statistics
##
## Accuracy : 0.9137
## 95% CI : (0.9047, 0.9222)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8959
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.9366 0.8401
## Specificity 0.99376 1.00000 0.9916 0.9821
## Pos Pred Value 0.94057 1.00000 0.9385 0.9430
## Neg Pred Value 0.99133 1.00000 0.9914 0.9457
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.03824 0.1123 0.2189
## Detection Prevalence 0.09485 0.03824 0.1196 0.2321
## Balanced Accuracy 0.95647 1.00000 0.9641 0.9111
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9567 0.9474 0.9215
## Specificity 0.9934 0.9905 0.9426
## Pos Pred Value 0.9601 0.9458 0.7939
## Neg Pred Value 0.9929 0.9908 0.9804
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1355 0.1412 0.1784
## Detection Prevalence 0.1412 0.1493 0.2248
## Balanced Accuracy 0.9751 0.9689 0.9320
nb_tda_kde_5.50.5_n1_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9137255 0.8958994 0.9046885 0.9221647 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
nb_tda_kde_5.50.5_n1_db_lr_cf0_ov_acc<-nb_tda_kde_5.50.5_n1_db_lr_cf0$overall[1]
nb_tda_kde_5.50.5_n1_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.9937568 0.9405685 0.9913350 0.9405685
## Class: BOMBAY 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
## Class: CALI 0.9366053 0.9916458 0.9385246 0.9913697 0.9385246
## Class: DERMASON 0.8400753 0.9821014 0.9429778 0.9457389 0.9429778
## Class: HOROZ 0.9567474 0.9934323 0.9600694 0.9928653 0.9600694
## Class: SEKER 0.9473684 0.9904954 0.9458128 0.9907808 0.9458128
## Class: SIRA 0.9215190 0.9425532 0.7938931 0.9803984 0.7938931
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.9297573 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 1.0000000 0.03823529 0.03823529
## Class: CALI 0.9366053 0.9375640 0.11985294 0.11225490
## Class: DERMASON 0.8400753 0.8885572 0.26053922 0.21887255
## Class: HOROZ 0.9567474 0.9584055 0.14166667 0.13553922
## Class: SEKER 0.9473684 0.9465900 0.14901961 0.14117647
## Class: SIRA 0.9215190 0.8529584 0.19362745 0.17843137
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09485294 0.9564744
## Class: BOMBAY 0.03823529 1.0000000
## Class: CALI 0.11960784 0.9641255
## Class: DERMASON 0.23210784 0.9110883
## Class: HOROZ 0.14117647 0.9750899
## Class: SEKER 0.14926471 0.9689319
## Class: SIRA 0.22475490 0.9320361
nb_tda_kde_5.50.5_n1_db_lr_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n1_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_lr_n1_3_fold<-(db_lr_fit_re - nb_tda_kde_5.50.5_n1_lr_fit_re)
diff_drybean_tda_kde_5.50.5_lr_n1_3_fold
## Accuracy
## 1 -0.01546042
## 2 -0.03360840
## 3 -0.03010587
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n1_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n1_3_fold_odds.left<-bst_tda_kde_5.50.5_lr.n1_3_fold$probLeft/bst_tda_kde_5.50.5_lr.n1_3_fold$probRight
bst_tda_kde_5.50.5_lr.n1_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n1_3_fold
## $winLeft
## [1] 0.9617667
##
## $winRope
## [1] 0.03823333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n1_3_fold
## $left
## [1] 0.9374152
##
## $rope
## [1] 0.04772251
##
## $right
## [1] 0.01486228
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold))
#bf_tda_kde_5.50.5_lr.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_lr_n1_3_fold)
## t = -4.7481, df = 2, p-value = 0.04161
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.050307031 -0.002476099
## sample estimates:
## mean of x
## -0.02639156
### Test set diff
diff_drybean_tda_kde_5.50.5_lr.n1_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n1_db_lr_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_lr.n1_test
## Accuracy
## 0.01323529
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n1_test),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n1_test_odds.left<-bst_tda_kde_5.50.5_lr.n1_test$probLeft/bst_tda_kde_5.50.5_lr.n1_test$probRight
bst_tda_kde_5.50.5_lr.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n1_test),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.4588667
##
## $winRight
## [1] 0.5411333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n1_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_lr.n1_test)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr.n1_test)) #bf_tda_pca_5.50.5_lr.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n1_test))
##With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node2
DryBean_TDA_KDE_5.50.5_n2_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_kde_dry_bean_dataset_5.50.5.n2.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4841.754969
## iter 20 value 3613.144007
## iter 30 value 2045.358066
## iter 40 value 956.594837
## iter 50 value 904.920532
## iter 60 value 879.362864
## iter 70 value 869.733222
## iter 80 value 859.111981
## iter 90 value 852.150184
## iter 100 value 846.790046
## final value 846.790046
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4841.755378
## iter 20 value 3613.149625
## iter 30 value 2198.848825
## iter 40 value 1099.427381
## iter 50 value 976.483536
## iter 60 value 951.343313
## iter 70 value 939.587329
## iter 80 value 937.307777
## iter 90 value 935.895627
## iter 100 value 935.701305
## final value 935.701305
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4841.754969
## iter 20 value 3613.144047
## iter 30 value 2045.473504
## iter 40 value 957.004476
## iter 50 value 905.809522
## iter 60 value 881.449094
## iter 70 value 872.726143
## iter 80 value 863.761529
## iter 90 value 858.221546
## iter 100 value 854.288928
## final value 854.288928
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9059.135876
## iter 10 value 4797.864962
## iter 20 value 3239.933230
## iter 30 value 2246.095003
## iter 40 value 868.319317
## iter 50 value 813.916795
## iter 60 value 790.717458
## iter 70 value 781.747402
## iter 80 value 777.729583
## iter 90 value 771.840791
## iter 100 value 769.585146
## final value 769.585146
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9059.135876
## iter 10 value 4797.865380
## iter 20 value 3239.941531
## iter 30 value 1863.032982
## iter 40 value 996.341382
## iter 50 value 909.554614
## iter 60 value 885.764824
## iter 70 value 874.365007
## iter 80 value 872.518879
## iter 90 value 871.941056
## iter 100 value 871.876120
## final value 871.876120
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9059.135876
## iter 10 value 4797.864962
## iter 20 value 3239.933263
## iter 30 value 2246.223632
## iter 40 value 868.859521
## iter 50 value 815.125963
## iter 60 value 793.398271
## iter 70 value 785.501274
## iter 80 value 782.139346
## iter 90 value 777.437977
## iter 100 value 775.744918
## final value 775.744918
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4885.533572
## iter 20 value 3254.751513
## iter 30 value 1591.150344
## iter 40 value 892.885881
## iter 50 value 861.906289
## iter 60 value 843.176159
## iter 70 value 835.597137
## iter 80 value 828.356986
## iter 90 value 822.923143
## iter 100 value 818.553246
## final value 818.553246
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4885.534002
## iter 20 value 3254.770451
## iter 30 value 1662.261780
## iter 40 value 1034.946397
## iter 50 value 947.479698
## iter 60 value 923.997427
## iter 70 value 913.141775
## iter 80 value 911.564098
## iter 90 value 911.190393
## iter 100 value 911.115252
## final value 911.115252
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 9055.552357
## iter 10 value 4885.533572
## iter 20 value 3254.751552
## iter 30 value 1591.227782
## iter 40 value 893.443994
## iter 50 value 862.961428
## iter 60 value 845.394692
## iter 70 value 838.641043
## iter 80 value 832.569662
## iter 90 value 828.114443
## iter 100 value 824.957455
## final value 824.957455
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 13585.120296
## iter 10 value 7273.804421
## iter 20 value 4067.771438
## iter 30 value 2518.237578
## iter 40 value 1337.595482
## iter 50 value 1286.454409
## iter 60 value 1262.308765
## iter 70 value 1249.188251
## iter 80 value 1243.729097
## iter 90 value 1238.157229
## iter 100 value 1233.653115
## final value 1233.653115
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n2_LrFit0
## Penalized Multinomial Regression
##
## 7582 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5054, 5056, 5054
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9460565 0.9310555
## 1e-04 0.9465841 0.9317214
## 1e-01 0.9423631 0.9263261
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_KDE_5.50.5_n2_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9469517 0.9322544 Fold2
## 2 0.9465981 0.9317144 Fold1
## 3 0.9462025 0.9311955 Fold3
nb_tda_kde_5.50.5_n2_lr_fit_re<-DryBean_TDA_KDE_5.50.5_n2_LrFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n2_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI 19.05626 0.002972756 -0.18443391 2.37346646 2.8822418
## DERMASON 19.05755 0.001097468 -0.04111903 1.44714806 1.4100183
## HOROZ -10.81143 0.007910822 0.04351992 2.18621975 4.6380524
## SEKER -10.40457 0.004020212 0.09668340 -0.09602789 -0.6554541
## SIRA 57.72405 0.003361244 -0.43402571 2.23762796 2.9161683
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI -52.31975 32.05919 -0.003380958 -4.4444531 5.786511
## DERMASON -33.44919 23.04210 -0.002968513 -2.5642276 -15.821565
## HOROZ 47.49808 67.61948 -0.006171255 -7.4157268 -5.292679
## SEKER 13.55370 -80.56346 -0.003778468 0.1267185 -10.843799
## SIRA -42.48669 101.05967 -0.004301807 -3.5098273 -8.516168
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI 20.81700 -43.59746 12.48273 0.31568270 0.124237561
## DERMASON 3.88883 37.49429 14.10902 0.09331638 0.008089463
## HOROZ 34.95519 51.96063 -36.60310 0.06427325 -0.404167894
## SEKER -23.84906 85.71036 13.52889 -0.28296050 0.176556626
## SIRA 29.19704 -159.14051 39.83074 1.05375697 0.348855747
## ShapeFactor3 ShapeFactor4
## CALI 2.410771 5.178713
## DERMASON 1.157556 7.848230
## HOROZ -56.880568 -6.868662
## SEKER 43.828024 7.621281
## SIRA 7.794554 20.303658
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI 4.128369e-06 0.0005467353 0.001833236 0.0008854458 0.0006586432
## DERMASON 9.550503e-06 0.0009669833 0.003205864 0.0018168526 0.0020248641
## HOROZ 4.593080e-06 0.0006432548 0.002013569 0.0007996313 0.0008381670
## SEKER 8.003805e-06 0.0012080434 0.003023476 0.0006288964 0.0014312720
## SIRA 8.267572e-06 0.0006674013 0.002573272 0.0034336249 0.0034765827
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI 7.901234e-06 3.195765e-06 0.0005438585 0.0005760083 3.370255e-06
## DERMASON 2.059419e-05 7.785112e-06 0.0009819891 0.0010286804 9.073847e-06
## HOROZ 6.997347e-06 3.005486e-06 0.0006384838 0.0006917930 4.184887e-06
## SEKER 6.417475e-06 3.000065e-06 0.0012109828 0.0010104162 6.925234e-06
## SIRA 3.648878e-05 1.112685e-05 0.0006716057 0.0010123793 9.787766e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI 4.075204e-06 4.108339e-06 3.871547e-06 3.474644e-08 1.297776e-08
## DERMASON 9.431160e-06 1.012105e-05 1.269471e-05 8.634762e-08 6.658832e-08
## HOROZ 4.504002e-06 4.718564e-06 4.658357e-06 3.408746e-08 1.570244e-08
## SEKER 7.925962e-06 7.911783e-06 8.838574e-06 5.982171e-08 3.533132e-08
## SIRA 8.143670e-06 1.349461e-05 1.756611e-05 5.156651e-08 9.225182e-08
## ShapeFactor3 ShapeFactor4
## CALI 3.803370e-06 4.101324e-06
## DERMASON 1.509421e-05 9.554445e-06
## HOROZ 4.591064e-06 4.524234e-06
## SEKER 9.207185e-06 8.000819e-06
## SIRA 2.315503e-05 8.314071e-06
##
## Residual Deviance: 2467.306
## AIC: 2637.306
vip(DryBean_TDA_KDE_5.50.5_n2_LrFit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n2_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n2_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n2_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n2_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n2_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 15 1 1 7 5
## BOMBAY 0 0 0 0 0 0 0
## CALI 19 0 461 0 18 2 1
## DERMASON 0 0 0 946 6 7 74
## HOROZ 3 156 6 5 543 0 7
## SEKER 2 0 1 24 0 574 2
## SIRA 8 0 6 87 10 18 701
##
## Overall Statistics
##
## Accuracy : 0.8797
## 95% CI : (0.8693, 0.8895)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8539
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 0.00000 0.9427 0.8899
## Specificity 0.99213 1.00000 0.9889 0.9712
## Pos Pred Value 0.92621 NaN 0.9202 0.9158
## Neg Pred Value 0.99132 0.96176 0.9922 0.9616
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.00000 0.1130 0.2319
## Detection Prevalence 0.09632 0.00000 0.1228 0.2532
## Balanced Accuracy 0.95566 0.50000 0.9658 0.9305
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9394 0.9441 0.8873
## Specificity 0.9495 0.9916 0.9608
## Pos Pred Value 0.7542 0.9519 0.8446
## Neg Pred Value 0.9896 0.9902 0.9726
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1331 0.1407 0.1718
## Detection Prevalence 0.1765 0.1478 0.2034
## Balanced Accuracy 0.9445 0.9679 0.9241
nb_tda_kde_5.50.5_n2_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 15 1 1 7 5
## BOMBAY 0 0 0 0 0 0 0
## CALI 19 0 461 0 18 2 1
## DERMASON 0 0 0 946 6 7 74
## HOROZ 3 156 6 5 543 0 7
## SEKER 2 0 1 24 0 574 2
## SIRA 8 0 6 87 10 18 701
##
## Overall Statistics
##
## Accuracy : 0.8797
## 95% CI : (0.8693, 0.8895)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8539
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 0.00000 0.9427 0.8899
## Specificity 0.99213 1.00000 0.9889 0.9712
## Pos Pred Value 0.92621 NaN 0.9202 0.9158
## Neg Pred Value 0.99132 0.96176 0.9922 0.9616
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.08922 0.00000 0.1130 0.2319
## Detection Prevalence 0.09632 0.00000 0.1228 0.2532
## Balanced Accuracy 0.95566 0.50000 0.9658 0.9305
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9394 0.9441 0.8873
## Specificity 0.9495 0.9916 0.9608
## Pos Pred Value 0.7542 0.9519 0.8446
## Neg Pred Value 0.9896 0.9902 0.9726
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1331 0.1407 0.1718
## Detection Prevalence 0.1765 0.1478 0.2034
## Balanced Accuracy 0.9445 0.9679 0.9241
nb_tda_kde_5.50.5_n2_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8796569 0.8538735 0.8692781 0.8894885 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
nb_tda_kde_5.50.5_n2_db_lr_cf0_ov_acc<-nb_tda_kde_5.50.5_n2_db_lr_cf0$overall[1]
nb_tda_kde_5.50.5_n2_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.9921281 0.9262087 0.9913209 0.9262087
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.9427403 0.9888610 0.9201597 0.9921766 0.9201597
## Class: DERMASON 0.8899341 0.9711634 0.9157793 0.9616016 0.9157793
## Class: HOROZ 0.9394464 0.9494575 0.7541667 0.9895833 0.7541667
## Class: SEKER 0.9440789 0.9916475 0.9519071 0.9902215 0.9519071
## Class: SIRA 0.8873418 0.9607903 0.8445783 0.9726154 0.8445783
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.9226869 0.09705882 0.08921569
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.9427403 0.9313131 0.11985294 0.11299020
## Class: DERMASON 0.8899341 0.9026718 0.26053922 0.23186275
## Class: HOROZ 0.9394464 0.8366718 0.14166667 0.13308824
## Class: SEKER 0.9440789 0.9479769 0.14901961 0.14068627
## Class: SIRA 0.8873418 0.8654321 0.19362745 0.17181373
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09632353 0.9556600
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.12279412 0.9658007
## Class: DERMASON 0.25318627 0.9305488
## Class: HOROZ 0.17647059 0.9444519
## Class: SEKER 0.14779412 0.9678632
## Class: SIRA 0.20343137 0.9240660
nb_tda_kde_5.50.5_n2_db_lr_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n2_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_lr_n2_3_fold<-(db_lr_fit_re - nb_tda_kde_5.50.5_n2_lr_fit_re)
diff_drybean_tda_kde_5.50.5_lr_n2_3_fold
## Accuracy
## 1 -0.01055372
## 2 -0.02620792
## 3 -0.02237502
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n2_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n2_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n2_3_fold_odds.left<-bst_tda_kde_5.50.5_lr.n2_3_fold$probLeft/bst_tda_kde_5.50.5_lr.n2_3_fold$probRight
bst_tda_kde_5.50.5_lr.n2_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n2_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n2_3_fold
## $winLeft
## [1] 0.9654
##
## $winRope
## [1] 0.0346
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n2_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n2_3_fold
## $left
## [1] 0.8919397
##
## $rope
## [1] 0.0920987
##
## $right
## [1] 0.01596165
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold))
#bf_tda_kde_5.50.5_lr.n2_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_lr_n2_3_fold)
## t = -4.1843, df = 2, p-value = 0.05265
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.039982142 0.000557704
## sample estimates:
## mean of x
## -0.01971222
### Test set diff
diff_drybean_tda_kde_5.50.5_lr.n2_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n2_db_lr_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_lr.n2_test
## Accuracy
## 0.04730392
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n2_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n2_test_odds.left<-bst_tda_kde_5.50.5_lr.n2_test$probLeft/bst_tda_kde_5.50.5_lr.n2_test$probRight
bst_tda_kde_5.50.5_lr.n2_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1594333
##
## $winRight
## [1] 0.8405667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n2_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n2_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_lr.n2_test)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test)) #bf_tda_pca_5.50.5_lr.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test))
##Node3
DryBean_TDA_KDE_5.50.5_n3_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_kde_dry_bean_dataset_5.50.5.n3.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1195.706177
## iter 20 value 1065.011946
## iter 30 value 814.441109
## iter 40 value 677.658815
## iter 50 value 662.806446
## iter 60 value 648.184375
## iter 70 value 643.981421
## iter 80 value 639.639125
## iter 90 value 638.095403
## iter 100 value 636.750714
## final value 636.750714
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1195.707226
## iter 20 value 1065.019948
## iter 30 value 814.939777
## iter 40 value 697.797267
## iter 50 value 689.623825
## iter 60 value 686.109002
## iter 70 value 685.915664
## iter 80 value 685.845849
## iter 90 value 685.759311
## iter 100 value 685.737093
## final value 685.737093
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1195.706178
## iter 20 value 1065.011956
## iter 30 value 814.438714
## iter 40 value 677.854365
## iter 50 value 663.568928
## iter 60 value 650.647272
## iter 70 value 647.468128
## iter 80 value 644.450875
## iter 90 value 643.537196
## iter 100 value 642.575933
## final value 642.575933
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1288.032700
## iter 20 value 1181.673467
## iter 30 value 852.243571
## iter 40 value 647.917367
## iter 50 value 627.336754
## iter 60 value 610.211801
## iter 70 value 605.181875
## iter 80 value 598.132085
## iter 90 value 595.245127
## iter 100 value 593.634561
## final value 593.634561
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1288.034040
## iter 20 value 1181.682717
## iter 30 value 854.522248
## iter 40 value 666.936505
## iter 50 value 656.067444
## iter 60 value 654.607663
## iter 70 value 654.250515
## iter 80 value 654.220792
## iter 90 value 654.214878
## iter 100 value 654.212551
## final value 654.212551
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1288.032701
## iter 20 value 1181.673473
## iter 30 value 852.244633
## iter 40 value 648.056982
## iter 50 value 628.026888
## iter 60 value 612.862070
## iter 70 value 608.774775
## iter 80 value 603.982874
## iter 90 value 602.105647
## iter 100 value 601.205908
## final value 601.205908
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1224.952831
## iter 20 value 1040.910707
## iter 30 value 852.827231
## iter 40 value 706.297630
## iter 50 value 687.906681
## iter 60 value 673.832971
## iter 70 value 667.871553
## iter 80 value 664.580538
## iter 90 value 663.005357
## iter 100 value 661.798833
## final value 661.798833
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1224.954207
## iter 20 value 1040.918731
## iter 30 value 857.064042
## iter 40 value 732.568445
## iter 50 value 724.379975
## iter 60 value 722.151524
## iter 70 value 722.004282
## iter 80 value 721.989733
## iter 90 value 721.984724
## final value 721.984035
## converged
## # weights: 108 (85 variable)
## initial value 4956.006692
## iter 10 value 1224.952832
## iter 20 value 1040.910714
## iter 30 value 852.832432
## iter 40 value 706.522106
## iter 50 value 688.838648
## iter 60 value 676.257754
## iter 70 value 671.570896
## iter 80 value 669.340353
## iter 90 value 668.315497
## iter 100 value 667.567329
## final value 667.567329
## stopped after 100 iterations
## # weights: 108 (85 variable)
## initial value 7434.010038
## iter 10 value 1744.375409
## iter 20 value 1457.282547
## iter 30 value 1202.391812
## iter 40 value 1019.568915
## iter 50 value 992.648147
## iter 60 value 975.688104
## iter 70 value 968.729438
## iter 80 value 964.879512
## iter 90 value 963.568311
## iter 100 value 962.064725
## final value 962.064725
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n3_LrFit0
## Penalized Multinomial Regression
##
## 4149 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 2766, 2766, 2766
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.9156423 0.8726380
## 1e-04 0.9166064 0.8740869
## 1e-01 0.9103398 0.8646375
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_KDE_5.50.5_n3_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.9016631 0.8515150 Fold2
## 2 0.9219089 0.8820536 Fold1
## 3 0.9262473 0.8886922 Fold3
nb_tda_kde_5.50.5_n3_lr_fit_re<-DryBean_TDA_KDE_5.50.5_n2_LrFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n3_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI -0.4547832 -0.006686471 -0.39806753 0.3612705 0.2612429
## DERMASON -16.7215241 0.006064208 -0.05971107 1.3504345 2.1952024
## HOROZ 5.6639707 0.013078735 -0.03916328 2.9179190 4.4775828
## SEKER -11.9129808 0.017104902 -0.01422559 -0.4950714 -2.3118436
## SIRA 29.3743719 0.008363490 -0.19841544 2.2305729 3.5932557
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI -2.451518 2.900569 0.006791898 0.6637701 -2.842236
## DERMASON 13.706092 9.165755 -0.010656888 -2.1505813 -26.463131
## HOROZ 16.749448 25.632926 -0.014498033 -7.0049110 -28.027500
## SEKER 1.381761 -159.864136 -0.019912647 3.3416649 -25.055253
## SIRA -24.996908 107.326608 -0.009655952 -4.8362221 -18.055560
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI -0.5297753 -1.523597 -1.45190263 -0.002323848 -0.01460178
## DERMASON -29.6193325 31.511427 -17.38274531 -0.192787067 -0.15238350
## HOROZ 8.7620460 11.671619 -2.09244038 0.195824941 -0.09936350
## SEKER -13.2561450 8.974547 33.31830763 -0.565351671 0.53249528
## SIRA 37.1121303 -55.086724 -0.08419269 0.614973150 -0.13565132
## ShapeFactor3 ShapeFactor4
## CALI -2.640433 -1.374212
## DERMASON -17.956351 -14.237086
## HOROZ -9.316437 3.918728
## SEKER 84.506756 17.114260
## SIRA -35.926305 4.085741
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## CALI 1.232850e-07 0.001457998 3.591229e-05 0.0000259726 3.868781e-05
## DERMASON 8.294525e-06 0.001774921 1.753566e-03 0.0031355712 3.219529e-03
## HOROZ 9.016115e-06 0.002046382 3.490464e-03 0.0012307570 8.479790e-04
## SEKER 7.967033e-06 0.002212789 2.624260e-03 0.0007050524 1.007435e-03
## SIRA 8.679085e-06 0.001689665 1.796470e-03 0.0030693032 3.321951e-03
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## CALI 2.536659e-07 6.768354e-08 0.001444405 1.832427e-05 8.329950e-08
## DERMASON 3.420737e-05 1.243350e-05 0.001748342 8.880873e-04 8.903549e-06
## HOROZ 1.399760e-05 6.937918e-06 0.002009333 1.022886e-03 6.659905e-06
## SEKER 7.527676e-06 3.972568e-06 0.002190989 8.922566e-04 6.720800e-06
## SIRA 3.356080e-05 1.159736e-05 0.001663321 9.581201e-04 9.467200e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## CALI 1.758341e-07 2.972853e-07 2.238032e-07 3.540158e-10 1.068808e-09
## DERMASON 8.178422e-06 1.229758e-05 1.768299e-05 5.753505e-08 1.034629e-07
## HOROZ 8.862901e-06 7.702615e-06 7.193116e-06 9.638475e-08 2.317920e-08
## SEKER 7.905902e-06 8.242242e-06 7.827483e-06 7.213453e-08 3.345469e-08
## SIRA 8.565068e-06 1.264722e-05 1.824873e-05 4.898856e-08 1.041735e-07
## ShapeFactor3 ShapeFactor4
## CALI 2.808193e-07 1.562857e-07
## DERMASON 2.374322e-05 8.305578e-06
## HOROZ 5.732084e-06 8.956815e-06
## SEKER 7.539217e-06 7.961519e-06
## SIRA 2.420716e-05 8.727594e-06
##
## Residual Deviance: 1924.129
## AIC: 2094.129
vip(DryBean_TDA_KDE_5.50.5_n3_LrFit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n3_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n3_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n3_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n3_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n3_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 386 2 220 1 30 13 16
## BOMBAY 0 0 0 0 0 0 0
## CALI 6 154 262 0 2 14 7
## DERMASON 0 0 0 967 5 11 95
## HOROZ 0 0 2 5 514 0 11
## SEKER 0 0 0 25 0 560 2
## SIRA 4 0 5 65 27 10 659
##
## Overall Statistics
##
## Accuracy : 0.8206
## 95% CI : (0.8085, 0.8323)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.7827
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.97475 0.00000 0.53579 0.9097
## Specificity 0.92345 1.00000 0.94904 0.9632
## Pos Pred Value 0.57784 NaN 0.58876 0.8970
## Neg Pred Value 0.99707 0.96176 0.93755 0.9680
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.09461 0.00000 0.06422 0.2370
## Detection Prevalence 0.16373 0.00000 0.10907 0.2642
## Balanced Accuracy 0.94910 0.50000 0.74241 0.9364
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.8893 0.9211 0.8342
## Specificity 0.9949 0.9922 0.9663
## Pos Pred Value 0.9662 0.9540 0.8558
## Neg Pred Value 0.9820 0.9863 0.9604
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1260 0.1373 0.1615
## Detection Prevalence 0.1304 0.1439 0.1887
## Balanced Accuracy 0.9421 0.9566 0.9002
nb_tda_kde_5.50.5_n3_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 386 2 220 1 30 13 16
## BOMBAY 0 0 0 0 0 0 0
## CALI 6 154 262 0 2 14 7
## DERMASON 0 0 0 967 5 11 95
## HOROZ 0 0 2 5 514 0 11
## SEKER 0 0 0 25 0 560 2
## SIRA 4 0 5 65 27 10 659
##
## Overall Statistics
##
## Accuracy : 0.8206
## 95% CI : (0.8085, 0.8323)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.7827
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.97475 0.00000 0.53579 0.9097
## Specificity 0.92345 1.00000 0.94904 0.9632
## Pos Pred Value 0.57784 NaN 0.58876 0.8970
## Neg Pred Value 0.99707 0.96176 0.93755 0.9680
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.09461 0.00000 0.06422 0.2370
## Detection Prevalence 0.16373 0.00000 0.10907 0.2642
## Balanced Accuracy 0.94910 0.50000 0.74241 0.9364
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.8893 0.9211 0.8342
## Specificity 0.9949 0.9922 0.9663
## Pos Pred Value 0.9662 0.9540 0.8558
## Neg Pred Value 0.9820 0.9863 0.9604
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1260 0.1373 0.1615
## Detection Prevalence 0.1304 0.1439 0.1887
## Balanced Accuracy 0.9421 0.9566 0.9002
nb_tda_kde_5.50.5_n3_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8205882 0.7827271 0.8084640 0.8322516 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
nb_tda_kde_5.50.5_n3_db_lr_cf0_ov_acc<-nb_tda_kde_5.50.5_n3_db_lr_cf0$overall[1]
nb_tda_kde_5.50.5_n3_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9747475 0.9234528 0.5778443 0.9970692 0.5778443
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.5357873 0.9490393 0.5887640 0.9375516 0.5887640
## Class: DERMASON 0.9096896 0.9632085 0.8970315 0.9680213 0.8970315
## Class: HOROZ 0.8892734 0.9948601 0.9661654 0.9819617 0.9661654
## Class: SEKER 0.9210526 0.9922235 0.9540034 0.9862582 0.9540034
## Class: SIRA 0.8341772 0.9662614 0.8558442 0.9604230 0.8558442
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9747475 0.7255639 0.09705882 0.09460784
## Class: BOMBAY 0.0000000 NA 0.03823529 0.00000000
## Class: CALI 0.5357873 0.5610278 0.11985294 0.06421569
## Class: DERMASON 0.9096896 0.9033162 0.26053922 0.23700980
## Class: HOROZ 0.8892734 0.9261261 0.14166667 0.12598039
## Class: SEKER 0.9210526 0.9372385 0.14901961 0.13725490
## Class: SIRA 0.8341772 0.8448718 0.19362745 0.16151961
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.1637255 0.9491001
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.1090686 0.7424133
## Class: DERMASON 0.2642157 0.9364490
## Class: HOROZ 0.1303922 0.9420667
## Class: SEKER 0.1438725 0.9566381
## Class: SIRA 0.1887255 0.9002193
nb_tda_kde_5.50.5_n3_db_lr_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n3_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_lr_n3_3_fold<-(db_lr_fit_re - nb_tda_kde_5.50.5_n3_lr_fit_re)
diff_drybean_tda_kde_5.50.5_lr_n3_3_fold
## Accuracy
## 1 -0.01055372
## 2 -0.02620792
## 3 -0.02237502
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n3_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n3_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n3_3_fold_odds.left<-bst_tda_kde_5.50.5_lr.n3_3_fold$probLeft/bst_tda_kde_5.50.5_lr.n3_3_fold$probRight
bst_tda_kde_5.50.5_lr.n3_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n3_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n3_3_fold
## $winLeft
## [1] 0.9626667
##
## $winRope
## [1] 0.03733333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n3_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n3_3_fold
## $left
## [1] 0.8919397
##
## $rope
## [1] 0.0920987
##
## $right
## [1] 0.01596165
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold))
#bf_tda_kde_5.50.5_lr.n3_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_lr_n3_3_fold)
## t = -4.1843, df = 2, p-value = 0.05265
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.039982142 0.000557704
## sample estimates:
## mean of x
## -0.01971222
### Test set diff
diff_drybean_tda_kde_5.50.5_lr.n3_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n3_db_lr_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_lr.n3_test
## Accuracy
## 0.1063725
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n3_test),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n3_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n3_test_odds.left<-bst_tda_kde_5.50.5_lr.n3_test$probLeft/bst_tda_kde_5.50.5_lr.n3_test$probRight
bst_tda_kde_5.50.5_lr.n3_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n2_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n2_test),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n2_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1600667
##
## $winRight
## [1] 0.8399333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n3_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n3_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n3_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_lr.n3_test)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr.n3_test)) #bf_tda_pca_5.50.5_lr.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n3_test))
##Node4
DryBean_TDA_KDE_5.50.5_n4_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_kde_dry_bean_dataset_5.50.5.n4.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 72 (51 variable)
## initial value 1871.497388
## iter 10 value 930.739283
## iter 20 value 662.968620
## iter 30 value 562.949721
## iter 40 value 558.699683
## iter 50 value 555.577200
## iter 60 value 554.718693
## iter 70 value 554.250782
## iter 80 value 552.902177
## iter 90 value 549.601719
## iter 100 value 548.240729
## final value 548.240729
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1871.497388
## iter 10 value 930.743291
## iter 20 value 654.196441
## iter 30 value 572.553835
## iter 40 value 571.378512
## iter 50 value 570.549830
## final value 570.549812
## converged
## # weights: 72 (51 variable)
## initial value 1871.497388
## iter 10 value 930.739290
## iter 20 value 662.979260
## iter 30 value 563.002734
## iter 40 value 558.960867
## iter 50 value 556.147439
## iter 60 value 555.467776
## iter 70 value 555.153613
## iter 80 value 554.295152
## iter 90 value 553.568951
## iter 100 value 553.453282
## final value 553.453282
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 1065.352490
## iter 20 value 638.738236
## iter 30 value 577.977780
## iter 40 value 569.061474
## iter 50 value 567.166938
## iter 60 value 565.350490
## iter 70 value 564.263113
## iter 80 value 562.412316
## iter 90 value 556.982392
## iter 100 value 554.327607
## final value 554.327607
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 1065.363706
## iter 20 value 641.510031
## iter 30 value 585.575220
## iter 40 value 579.250476
## iter 50 value 579.209985
## iter 50 value 579.209981
## iter 50 value 579.209981
## final value 579.209981
## converged
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 1065.352500
## iter 20 value 638.737255
## iter 30 value 578.023357
## iter 40 value 569.271829
## iter 50 value 567.480117
## iter 60 value 566.147805
## iter 70 value 565.410862
## iter 80 value 564.552597
## iter 90 value 563.281814
## iter 100 value 563.098932
## final value 563.098932
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 981.352959
## iter 20 value 624.524214
## iter 30 value 552.178909
## iter 40 value 544.090357
## iter 50 value 541.115949
## iter 60 value 540.345399
## iter 70 value 537.280546
## iter 80 value 533.336018
## iter 90 value 530.758399
## iter 100 value 529.299163
## final value 529.299163
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 981.356752
## iter 20 value 627.818755
## iter 30 value 563.383790
## iter 40 value 557.659687
## iter 50 value 557.157218
## final value 557.157147
## converged
## # weights: 72 (51 variable)
## initial value 1870.111093
## iter 10 value 981.352962
## iter 20 value 624.527971
## iter 30 value 552.262266
## iter 40 value 544.437907
## iter 50 value 541.613951
## iter 60 value 540.999614
## iter 70 value 539.130345
## iter 80 value 537.963976
## iter 90 value 537.341228
## iter 100 value 537.114039
## final value 537.114039
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 2805.859787
## iter 10 value 1225.844690
## iter 20 value 937.240180
## iter 30 value 843.229442
## iter 40 value 840.457792
## iter 50 value 836.485621
## iter 60 value 835.697501
## iter 70 value 833.187020
## iter 80 value 831.611471
## iter 90 value 831.283989
## iter 100 value 831.004772
## final value 831.004772
## stopped after 100 iterations
DryBean_TDA_KDE_5.50.5_n4_LrFit0
## Penalized Multinomial Regression
##
## 2024 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1350, 1349, 1349
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.8246049 0.7074089
## 1e-04 0.8265787 0.7107460
## 1e-01 0.8241089 0.7053588
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 1e-04.
DryBean_TDA_KDE_5.50.5_n4_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.8459259 0.7397359 Fold2
## 2 0.8219585 0.7073075 Fold1
## 3 0.8118519 0.6851945 Fold3
nb_tda_kde_5.50.5_n4_lr_fit_re<-DryBean_TDA_KDE_5.50.5_n4_LrFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n4_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## HOROZ 4.918146 0.002934240 0.0340530 1.2686575 1.890670
## SEKER 14.549214 0.016832642 0.2349197 -2.4124255 -4.111577
## SIRA -7.805785 -0.003940889 -0.2229275 0.4346715 1.073650
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## HOROZ 6.7976035 11.50951 -0.0014588191 -3.6236455 -2.121843
## SEKER -0.4335939 -98.54492 -0.0127610504 4.3973803 6.895134
## SIRA -20.8535497 45.79276 0.0001936836 0.7839925 8.060296
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2 ShapeFactor3
## HOROZ 3.3594006 2.073014 0.9943935 0.132559269 -0.01506593 -2.774977
## SEKER 9.3460751 29.410410 43.6604763 -0.001089737 0.57677371 74.855437
## SIRA -0.9901068 -129.867401 -22.4596521 -0.047209493 -0.28273544 -39.151457
## ShapeFactor4
## HOROZ 2.594087
## SEKER 28.878592
## SIRA -17.729635
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## HOROZ 6.143166e-07 0.0031303830 0.0004471674 0.0001467535 1.793773e-05
## SEKER 1.719594e-05 0.0018293913 0.0059398483 0.0018238268 1.923942e-03
## SIRA 6.269943e-06 0.0008268668 0.0025157936 0.0011446475 5.332347e-04
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## HOROZ 1.635039e-06 7.070129e-07 0.0030824329 5.833541e-05 2.617217e-07
## SEKER 1.926801e-05 9.285965e-06 0.0018674208 1.907210e-03 1.347836e-05
## SIRA 1.261138e-05 5.910277e-06 0.0008427697 6.885449e-04 4.678656e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## HOROZ 4.478714e-07 1.241902e-07 2.757438e-07 8.992689e-09 2.984888e-10
## SEKER 1.705822e-05 1.675000e-05 1.603274e-05 1.629743e-07 6.744617e-08
## SIRA 6.215288e-06 5.129600e-06 4.732800e-06 7.310132e-08 1.734507e-08
## ShapeFactor3 ShapeFactor4
## HOROZ 5.632570e-08 5.071539e-07
## SEKER 1.484440e-05 1.719834e-05
## SIRA 3.972577e-06 6.254176e-06
##
## Residual Deviance: 1662.01
## AIC: 1764.01
vip(DryBean_TDA_KDE_5.50.5_n4_LrFit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n4_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n4_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n4_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n4_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n4_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 1 0 0 958 3 10 92
## HOROZ 26 0 264 11 571 0 138
## SEKER 366 156 224 38 0 586 36
## SIRA 3 0 1 56 4 12 524
##
## Overall Statistics
##
## Accuracy : 0.6468
## 95% CI : (0.6319, 0.6615)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5678
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9012
## Specificity 1.00000 1.00000 1.0000 0.9649
## Pos Pred Value NaN NaN NaN 0.9004
## Neg Pred Value 0.90294 0.96176 0.8801 0.9652
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2348
## Detection Prevalence 0.00000 0.00000 0.0000 0.2608
## Balanced Accuracy 0.50000 0.50000 0.5000 0.9330
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9879 0.9638 0.6633
## Specificity 0.8746 0.7638 0.9769
## Pos Pred Value 0.5653 0.4168 0.8733
## Neg Pred Value 0.9977 0.9918 0.9236
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1400 0.1436 0.1284
## Detection Prevalence 0.2475 0.3446 0.1471
## Balanced Accuracy 0.9313 0.8638 0.8201
nb_tda_kde_5.50.5_n4_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 1 0 0 958 3 10 92
## HOROZ 26 0 264 11 571 0 138
## SEKER 366 156 224 38 0 586 36
## SIRA 3 0 1 56 4 12 524
##
## Overall Statistics
##
## Accuracy : 0.6468
## 95% CI : (0.6319, 0.6615)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5678
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.9012
## Specificity 1.00000 1.00000 1.0000 0.9649
## Pos Pred Value NaN NaN NaN 0.9004
## Neg Pred Value 0.90294 0.96176 0.8801 0.9652
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.2348
## Detection Prevalence 0.00000 0.00000 0.0000 0.2608
## Balanced Accuracy 0.50000 0.50000 0.5000 0.9330
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9879 0.9638 0.6633
## Specificity 0.8746 0.7638 0.9769
## Pos Pred Value 0.5653 0.4168 0.8733
## Neg Pred Value 0.9977 0.9918 0.9236
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1400 0.1436 0.1284
## Detection Prevalence 0.2475 0.3446 0.1471
## Balanced Accuracy 0.9313 0.8638 0.8201
nb_tda_kde_5.50.5_n4_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6468137 0.5677871 0.6319242 0.6614927 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
nb_tda_kde_5.50.5_n4_db_lr_cf0_ov_acc<-nb_tda_kde_5.50.5_n4_db_lr_cf0$overall[1]
nb_tda_kde_5.50.5_n4_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.0000000 1.0000000 NaN 0.9029412 NA
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647 NA
## Class: CALI 0.0000000 1.0000000 NaN 0.8801471 NA
## Class: DERMASON 0.9012230 0.9648658 0.9003759 0.9651857 0.9003759
## Class: HOROZ 0.9878893 0.8746431 0.5653465 0.9977199 0.5653465
## Class: SEKER 0.9638158 0.7638249 0.4167852 0.9917726 0.4167852
## Class: SIRA 0.6632911 0.9768997 0.8733333 0.9235632 0.8733333
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.0000000 NA 0.09705882 0.0000000
## Class: BOMBAY 0.0000000 NA 0.03823529 0.0000000
## Class: CALI 0.0000000 NA 0.11985294 0.0000000
## Class: DERMASON 0.9012230 0.9007992 0.26053922 0.2348039
## Class: HOROZ 0.9878893 0.7191436 0.14166667 0.1399510
## Class: SEKER 0.9638158 0.5819265 0.14901961 0.1436275
## Class: SIRA 0.6632911 0.7539568 0.19362745 0.1284314
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.0000000 0.5000000
## Class: BOMBAY 0.0000000 0.5000000
## Class: CALI 0.0000000 0.5000000
## Class: DERMASON 0.2607843 0.9330444
## Class: HOROZ 0.2475490 0.9312662
## Class: SEKER 0.3446078 0.8638203
## Class: SIRA 0.1470588 0.8200954
nb_tda_kde_5.50.5_n4_db_lr_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n4_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_lr_n4_3_fold<-(db_lr_fit_re - nb_tda_kde_5.50.5_n4_lr_fit_re)
diff_drybean_tda_kde_5.50.5_lr_n4_3_fold
## Accuracy
## 1 0.09047206
## 2 0.09843173
## 3 0.11197566
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n4_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n4_3_fold_odds.left<-bst_tda_kde_5.50.5_lr.n4_3_fold$probLeft/bst_tda_kde_5.50.5_lr.n4_3_fold$probRight
bst_tda_kde_5.50.5_lr.n4_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n4_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.009166667
##
## $winRight
## [1] 0.9908333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n4_3_fold
## $left
## [1] 0.002145385
##
## $rope
## [1] 0.001045579
##
## $right
## [1] 0.996809
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold))
#bf_tda_kde_5.50.5_lr.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_lr_n4_3_fold)
## t = 15.978, df = 2, p-value = 0.003894
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.07328566 0.12730063
## sample estimates:
## mean of x
## 0.1002931
### Test set diff
diff_drybean_tda_kde_5.50.5_lr.n4_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n4_db_lr_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_lr.n4_test
## Accuracy
## 0.2801471
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n4_test),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n4_test_odds.left<-bst_tda_kde_5.50.5_lr.n4_test$probLeft/bst_tda_kde_5.50.5_lr.n4_test$probRight
bst_tda_kde_5.50.5_lr.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1603667
##
## $winRight
## [1] 0.8396333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n4_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_lr.n4_test)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr.n4_test)) #bf_tda_pca_5.50.5_lr.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n4_test))
##Node5
DryBean_TDA_KDE_5.50.5_n5_LrFit0 <- train(as.factor(Class) ~ .,
data = tda.m_kde_dry_bean_dataset_5.50.5.n5.vec,
family = 'binomial',
method = 'multinom',
trControl = fitControl,
metric='Accuracy')
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 473.572062
## iter 20 value 360.487467
## iter 30 value 338.976113
## iter 40 value 336.934926
## iter 50 value 333.715995
## iter 60 value 332.450956
## iter 70 value 331.726934
## iter 80 value 327.180446
## iter 90 value 325.558314
## iter 100 value 325.266000
## final value 325.266000
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 473.575327
## iter 20 value 365.671145
## iter 30 value 352.893390
## iter 40 value 352.798874
## final value 352.797176
## converged
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 473.572066
## iter 20 value 360.493681
## iter 30 value 339.024071
## iter 40 value 337.096165
## iter 50 value 334.492069
## iter 60 value 333.668045
## iter 70 value 333.334205
## iter 80 value 332.496906
## iter 90 value 332.426894
## iter 100 value 332.400301
## final value 332.400301
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 914.954278
## iter 10 value 477.859914
## iter 20 value 376.091907
## iter 30 value 349.283488
## iter 40 value 348.892466
## iter 50 value 347.999198
## iter 60 value 347.757715
## iter 70 value 347.382708
## iter 80 value 347.025186
## iter 90 value 345.019888
## iter 100 value 343.968336
## final value 343.968336
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 914.954278
## iter 10 value 477.863710
## iter 20 value 375.813524
## iter 30 value 359.838955
## iter 40 value 359.335738
## final value 359.335369
## converged
## # weights: 72 (51 variable)
## initial value 914.954278
## iter 10 value 477.859917
## iter 20 value 376.101644
## iter 30 value 349.320089
## iter 40 value 348.948253
## iter 50 value 348.240602
## iter 60 value 348.089343
## iter 70 value 347.899391
## iter 80 value 347.790865
## iter 90 value 347.611423
## iter 100 value 347.588855
## final value 347.588855
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 478.335654
## iter 20 value 371.788064
## iter 30 value 363.890815
## iter 40 value 362.233731
## iter 50 value 360.031004
## iter 60 value 359.509521
## iter 70 value 356.949748
## iter 80 value 354.182153
## iter 90 value 353.332066
## iter 100 value 352.492464
## final value 352.492464
## stopped after 100 iterations
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 478.340207
## iter 20 value 377.227567
## iter 30 value 372.436927
## iter 40 value 372.345430
## final value 372.343912
## converged
## # weights: 72 (51 variable)
## initial value 913.567984
## iter 10 value 478.335659
## iter 20 value 371.798023
## iter 30 value 363.914342
## iter 40 value 362.404538
## iter 50 value 360.811794
## iter 60 value 360.526530
## iter 70 value 360.041338
## iter 80 value 359.999631
## iter 90 value 359.948485
## final value 359.925901
## converged
## # weights: 72 (51 variable)
## initial value 1371.045123
## iter 10 value 809.052677
## iter 20 value 618.093157
## iter 30 value 545.533603
## iter 40 value 543.074284
## final value 542.911194
## converged
DryBean_TDA_KDE_5.50.5_n5_LrFit0
## Penalized Multinomial Regression
##
## 989 samples
## 16 predictor
## 4 classes: 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 659, 660, 659
## Resampling results across tuning parameters:
##
## decay Accuracy Kappa
## 0e+00 0.7482331 0.5462197
## 1e-04 0.7492432 0.5488360
## 1e-01 0.7522796 0.5477844
##
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was decay = 0.1.
DryBean_TDA_KDE_5.50.5_n5_LrFit0$resample
## Accuracy Kappa Resample
## 1 0.7545455 0.5523963 Fold1
## 2 0.7454545 0.5384923 Fold3
## 3 0.7568389 0.5524647 Fold2
nb_tda_kde_5.50.5_n5_lr_fit_re<-DryBean_TDA_KDE_5.50.5_n5_LrFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n5_LrFit0)
## Call:
## nnet::multinom(formula = .outcome ~ ., data = dat, decay = param$decay,
## family = "binomial")
##
## Coefficients:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## HOROZ 0.007046354 -0.007017273 -0.009861693 -0.07326776 -0.7085975
## SEKER 0.039912415 0.019275103 0.210052507 -2.65315936 -2.8076537
## SIRA -0.058476682 0.002591046 0.105103512 0.69262706 1.0713977
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## HOROZ 0.01384804 0.01064355 0.005671292 0.8895502 0.1086813
## SEKER 0.03869226 -0.03413745 -0.017839900 4.6033418 0.3133238
## SIRA -0.39192565 0.30903918 -0.001628880 -2.2510964 4.6101021
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## HOROZ 0.007371226 0.01399459 0.008317498 3.231153e-05 2.904234e-05
## SEKER 0.039671461 0.08392669 0.080927850 -3.331845e-05 6.506517e-04
## SIRA -0.046568170 -0.16807132 -0.116932534 -1.159902e-03 -1.246201e-03
## ShapeFactor3 ShapeFactor4
## HOROZ 0.008120907 0.01861115
## SEKER 0.115165415 0.11418006
## SIRA -0.215281243 -0.17422578
##
## Std. Errors:
## (Intercept) Area Perimeter MajorAxisLength MinorAxisLength
## HOROZ 7.464852e-07 0.006744161 0.001017042 0.0001903871 1.168177e-06
## SEKER 2.380020e-05 0.002211257 0.010005601 0.0030792136 2.275544e-03
## SIRA 8.937963e-06 0.001099197 0.004150226 0.0023576757 6.086535e-04
## AspectRation Eccentricity ConvexArea EquivDiameter Extent
## HOROZ 2.242900e-06 9.582827e-07 0.006634756 6.042939e-05 9.410022e-07
## SEKER 3.240715e-05 1.602362e-05 0.002310454 2.642834e-03 1.646313e-05
## SIRA 2.615937e-05 1.227216e-05 0.001129689 9.824652e-04 5.717465e-06
## Solidity roundness Compactness ShapeFactor1 ShapeFactor2
## HOROZ 3.736457e-07 1.343551e-06 2.540216e-07 1.210208e-08 1.541372e-10
## SEKER 2.361202e-05 1.868319e-05 2.040759e-05 2.393448e-07 7.962231e-08
## SIRA 8.860390e-06 5.857470e-06 5.060119e-06 1.235992e-07 2.113535e-08
## ShapeFactor3 ShapeFactor4
## HOROZ 7.372696e-08 5.898251e-07
## SEKER 1.751545e-05 2.375758e-05
## SIRA 4.646547e-06 8.917319e-06
##
## Residual Deviance: 1085.822
## AIC: 1187.822
vip(DryBean_TDA_KDE_5.50.5_n5_LrFit0,50) + ggtitle("DryBean_TDA_KDE_5.50.5_n5_Lr1Fit TDA-Assited LR")
# Predict outcome using DryBean_TDA_KDE_5.50.5_n5_LrFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n5_LrFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n5_db_lr_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in levels(reference) != levels(data): Levels are not in the same order
## for reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n5_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 751 1 10 80
## HOROZ 0 0 0 213 10 0 1
## SEKER 68 68 1 16 0 576 6
## SIRA 328 88 488 83 567 22 703
##
## Overall Statistics
##
## Accuracy : 0.5
## 95% CI : (0.4845, 0.5155)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3777
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.7065
## Specificity 1.00000 1.00000 1.0000 0.9698
## Pos Pred Value NaN NaN NaN 0.8919
## Neg Pred Value 0.90294 0.96176 0.8801 0.9036
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.1841
## Detection Prevalence 0.00000 0.00000 0.0000 0.2064
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8382
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.017301 0.9474 0.8899
## Specificity 0.938892 0.9542 0.5210
## Pos Pred Value 0.044643 0.7837 0.3085
## Neg Pred Value 0.852697 0.9904 0.9517
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.002451 0.1412 0.1723
## Detection Prevalence 0.054902 0.1801 0.5586
## Balanced Accuracy 0.478097 0.9508 0.7054
nb_tda_kde_5.50.5_n5_db_lr_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 0 0 0 0 0 0 0
## BOMBAY 0 0 0 0 0 0 0
## CALI 0 0 0 0 0 0 0
## DERMASON 0 0 0 751 1 10 80
## HOROZ 0 0 0 213 10 0 1
## SEKER 68 68 1 16 0 576 6
## SIRA 328 88 488 83 567 22 703
##
## Overall Statistics
##
## Accuracy : 0.5
## 95% CI : (0.4845, 0.5155)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.3777
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.00000 0.00000 0.0000 0.7065
## Specificity 1.00000 1.00000 1.0000 0.9698
## Pos Pred Value NaN NaN NaN 0.8919
## Neg Pred Value 0.90294 0.96176 0.8801 0.9036
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.00000 0.00000 0.0000 0.1841
## Detection Prevalence 0.00000 0.00000 0.0000 0.2064
## Balanced Accuracy 0.50000 0.50000 0.5000 0.8382
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.017301 0.9474 0.8899
## Specificity 0.938892 0.9542 0.5210
## Pos Pred Value 0.044643 0.7837 0.3085
## Neg Pred Value 0.852697 0.9904 0.9517
## Prevalence 0.141667 0.1490 0.1936
## Detection Rate 0.002451 0.1412 0.1723
## Detection Prevalence 0.054902 0.1801 0.5586
## Balanced Accuracy 0.478097 0.9508 0.7054
nb_tda_kde_5.50.5_n5_db_lr_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.000000e-01 3.776857e-01 4.845399e-01 5.154601e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 2.853660e-233 NaN
nb_tda_kde_5.50.5_n5_db_lr_cf0_ov_acc<-nb_tda_kde_5.50.5_n5_db_lr_cf0$overall[1]
nb_tda_kde_5.50.5_n5_db_lr_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0.00000000 1.0000000 NaN 0.9029412
## Class: BOMBAY 0.00000000 1.0000000 NaN 0.9617647
## Class: CALI 0.00000000 1.0000000 NaN 0.8801471
## Class: DERMASON 0.70649106 0.9698376 0.89192399 0.9036442
## Class: HOROZ 0.01730104 0.9388921 0.04464286 0.8526971
## Class: SEKER 0.94736842 0.9542051 0.78367347 0.9904335
## Class: SIRA 0.88987342 0.5209726 0.30846863 0.9516935
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA NA 0.00000000 NA 0.09705882 0.00000000
## Class: BOMBAY NA 0.00000000 NA 0.03823529 0.00000000
## Class: CALI NA 0.00000000 NA 0.11985294 0.00000000
## Class: DERMASON 0.89192399 0.70649106 0.78845144 0.26053922 0.18406863
## Class: HOROZ 0.04464286 0.01730104 0.02493766 0.14166667 0.00245098
## Class: SEKER 0.78367347 0.94736842 0.85778109 0.14901961 0.14117647
## Class: SIRA 0.30846863 0.88987342 0.45812968 0.19362745 0.17230392
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.00000000 0.5000000
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.00000000 0.5000000
## Class: DERMASON 0.20637255 0.8381643
## Class: HOROZ 0.05490196 0.4780965
## Class: SEKER 0.18014706 0.9507867
## Class: SIRA 0.55857843 0.7054230
nb_tda_kde_5.50.5_n5_db_lr_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n5_db_lr_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_lr_n5_3_fold<-(db_lr_fit_re - nb_tda_kde_5.50.5_n5_lr_fit_re)
diff_drybean_tda_kde_5.50.5_lr_n5_3_fold
## Accuracy
## 1 0.1818525
## 2 0.1749356
## 3 0.1669886
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n5_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n5_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n5_3_fold_odds.left<-bst_tda_kde_5.50.5_lr.n5_3_fold$probLeft/bst_tda_kde_5.50.5_lr.n5_3_fold$probRight
bst_tda_kde_5.50.5_lr.n5_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n5_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n5_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.009966667
##
## $winRight
## [1] 0.9900333
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n5_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n5_3_fold
## $left
## [1] 0.0003604065
##
## $rope
## [1] 9.278293e-05
##
## $right
## [1] 0.9995468
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold))
#bf_tda_kde_5.50.5_lr.n5_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_lr_n5_3_fold)
## t = 40.657, df = 2, p-value = 0.0006044
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.1561155 0.1930691
## sample estimates:
## mean of x
## 0.1745923
### Test set diff
diff_drybean_tda_kde_5.50.5_lr.n5_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n5_db_lr_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_lr.n5_test
## Accuracy
## 0.4269608
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n5_test),-0.01,0.01)
bst_tda_kde_5.50.5_lr.n5_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_lr.n5_test_odds.left<-bst_tda_kde_5.50.5_lr.n5_test$probLeft/bst_tda_kde_5.50.5_lr.n5_test$probRight
bst_tda_kde_5.50.5_lr.n5_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_lr.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n5_test),-0.01,0.01)
bsr_tda_kde_5.50.5_lr.n5_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1575333
##
## $winRight
## [1] 0.8424667
# Bayesian Correlated Test
bct_tda_kde_5.50.5_lr.n5_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n5_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_lr.n5_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_lr.n5_test)))
#BayesFactor
#bf_tda_kde_5.50.5_lr.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_lr.n5_test)) #bf_tda_pca_5.50.5_lr.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_lr.n5_test))
#naiveBayes
dryBeanNbFit <- train(as.factor(Class) ~ ., data = Dry_Bean_DatasetTrain,
method = 'nb',
trControl = fitControl,
metric='Accuracy')
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 20
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 20
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 96
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 5
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 27
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 27
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 7
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 72
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 72
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2087
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2838
dryBeanNbFit
## Naive Bayes
##
## 9531 samples
## 16 predictor
## 7 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 6353, 6354, 6355
## Resampling results across tuning parameters:
##
## usekernel Accuracy Kappa
## FALSE 0.8986477 0.8775870
## TRUE 0.9019001 0.8814099
##
## Tuning parameter 'fL' was held constant at a value of 0
## Tuning
## parameter 'adjust' was held constant at a value of 1
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were fL = 0, usekernel = TRUE and adjust
## = 1.
dryBeanNbFit$resample
## Accuracy Kappa Resample
## 1 0.8967904 0.8752535 Fold1
## 2 0.9017941 0.8812123 Fold2
## 3 0.9071159 0.8877637 Fold3
db_nb_fit_re<-dryBeanNbFit$resample[1]
summary(dryBeanNbFit)
## Length Class Mode
## apriori 7 table numeric
## tables 16 -none- list
## levels 7 -none- character
## call 6 -none- call
## x 16 data.frame list
## usekernel 1 -none- logical
## varnames 16 -none- character
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 3 data.frame list
## obsLevels 7 -none- character
## param 0 -none- list
#varImp (dryBeanNbFit)
# Predict outcome using model from training data based on testing data
predictions <- predict(dryBeanNbFit, newdata= Dry_Bean_DatasetTest)
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 5
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 93
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3898
# Create confusion matrix to assess model fit/performance on test data
db_nb_cf<-confusionMatrix(data=predictions, as.factor(Dry_Bean_DatasetTest$Class))
db_nb_cf
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 319 0 42 0 2 5 6
## BOMBAY 1 156 0 0 0 0 0
## CALI 56 0 433 0 11 0 1
## DERMASON 0 0 0 940 6 6 76
## HOROZ 3 0 9 1 545 0 19
## SEKER 0 0 0 30 0 575 7
## SIRA 17 0 5 92 14 22 681
##
## Overall Statistics
##
## Accuracy : 0.8944
## 95% CI : (0.8845, 0.9036)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.8723
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.80556 1.00000 0.8855 0.8843
## Specificity 0.98507 0.99975 0.9811 0.9708
## Pos Pred Value 0.85294 0.99363 0.8643 0.9144
## Neg Pred Value 0.97922 1.00000 0.9844 0.9597
## Prevalence 0.09706 0.03824 0.1199 0.2605
## Detection Rate 0.07819 0.03824 0.1061 0.2304
## Detection Prevalence 0.09167 0.03848 0.1228 0.2520
## Balanced Accuracy 0.89531 0.99987 0.9333 0.9276
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9429 0.9457 0.8620
## Specificity 0.9909 0.9893 0.9544
## Pos Pred Value 0.9445 0.9395 0.8195
## Neg Pred Value 0.9906 0.9905 0.9665
## Prevalence 0.1417 0.1490 0.1936
## Detection Rate 0.1336 0.1409 0.1669
## Detection Prevalence 0.1414 0.1500 0.2037
## Balanced Accuracy 0.9669 0.9675 0.9082
db_nb_cf$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8943627 0.8723330 0.8845246 0.9036320 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_nb_cf_ov_acc<-db_nb_cf$overall[1]
db_nb_cf$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.8055556 0.9850706 0.8529412 0.9792229 0.8529412
## Class: BOMBAY 1.0000000 0.9997452 0.9936306 1.0000000 0.9936306
## Class: CALI 0.8854806 0.9810638 0.8642715 0.9843532 0.8642715
## Class: DERMASON 0.8842897 0.9708320 0.9143969 0.9596986 0.9143969
## Class: HOROZ 0.9429066 0.9908624 0.9445407 0.9905795 0.9445407
## Class: SEKER 0.9457237 0.9893433 0.9395425 0.9904844 0.9395425
## Class: SIRA 0.8620253 0.9544073 0.8194946 0.9664512 0.8194946
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.8055556 0.8285714 0.09705882 0.07818627
## Class: BOMBAY 1.0000000 0.9968051 0.03823529 0.03823529
## Class: CALI 0.8854806 0.8747475 0.11985294 0.10612745
## Class: DERMASON 0.8842897 0.8990913 0.26053922 0.23039216
## Class: HOROZ 0.9429066 0.9437229 0.14166667 0.13357843
## Class: SEKER 0.9457237 0.9426230 0.14901961 0.14093137
## Class: SIRA 0.8620253 0.8402221 0.19362745 0.16691176
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.09166667 0.8953131
## Class: BOMBAY 0.03848039 0.9998726
## Class: CALI 0.12279412 0.9332722
## Class: DERMASON 0.25196078 0.9275608
## Class: HOROZ 0.14142157 0.9668845
## Class: SEKER 0.15000000 0.9675335
## Class: SIRA 0.20367647 0.9082163
db_nb_cf_pre_rec_f1<-db_nb_cf$byClass[5:7]
##With TDA PCA filter 5 intervals, 50% overlap, 5 bins
##Node1
DryBean_TDA_PC_5.50.5_n1_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n1.vec,
method = 'nb',
trControl = fitControl,
metric='Accuracy')
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 13
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 67
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 67
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1080
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 7
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 32
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 7
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 32
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 9
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 9
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2458
DryBean_TDA_PC_5.50.5_n1_NbFit0
## Naive Bayes
##
## 7839 samples
## 16 predictor
## 6 classes: 'BARBUNYA', 'CALI', 'DERMASON', 'HOROZ', 'SEKER', 'SIRA'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 5227, 5225, 5226
## Resampling results across tuning parameters:
##
## usekernel Accuracy Kappa
## FALSE 0.8534248 0.783064
## TRUE 0.8585274 0.789569
##
## Tuning parameter 'fL' was held constant at a value of 0
## Tuning
## parameter 'adjust' was held constant at a value of 1
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were fL = 0, usekernel = TRUE and adjust
## = 1.
DryBean_TDA_PC_5.50.5_n1_NbFit0$resample
## Accuracy Kappa Resample
## 1 0.8526034 0.7812269 Fold1
## 2 0.8565417 0.7870815 Fold2
## 3 0.8664370 0.8003985 Fold3
db_tda_pc_5.50.5_n1_nb_fit_re<-DryBean_TDA_PC_5.50.5_n1_NbFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n1_NbFit0)
## Length Class Mode
## apriori 6 table numeric
## tables 16 -none- list
## levels 6 -none- character
## call 6 -none- call
## x 16 data.frame list
## usekernel 1 -none- logical
## varnames 16 -none- character
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 3 data.frame list
## obsLevels 6 -none- character
## param 0 -none- list
# Predict outcome using DryBean_TDA_PC_5.50.5_n1_NbFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n1_NbFit0, newdata= Dry_Bean_DatasetTest)
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 5
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 93
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1080
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1082
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1084
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1086
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1087
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1088
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1090
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1091
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1093
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1094
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1095
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1096
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1097
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1098
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2080
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2082
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2084
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2086
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2087
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2088
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2090
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2091
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2093
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2094
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2095
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2096
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2097
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2098
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3628
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n1_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n1_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 294 0 183 0 253 17 15
## BOMBAY 0 0 0 0 0 0 0
## CALI 36 78 29 0 209 9 9
## DERMASON 0 0 0 873 1 6 59
## HOROZ 60 78 274 72 108 0 39
## SEKER 0 0 0 30 0 562 6
## SIRA 6 0 3 88 7 14 662
##
## Overall Statistics
##
## Accuracy : 0.6196
## 95% CI : (0.6045, 0.6345)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5418
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.74242 0.00000 0.059305 0.8213
## Specificity 0.87296 1.00000 0.905040 0.9781
## Pos Pred Value 0.38583 NaN 0.078378 0.9297
## Neg Pred Value 0.96926 0.96176 0.876011 0.9395
## Prevalence 0.09706 0.03824 0.119853 0.2605
## Detection Rate 0.07206 0.00000 0.007108 0.2140
## Detection Prevalence 0.18676 0.00000 0.090686 0.2301
## Balanced Accuracy 0.80769 0.50000 0.482173 0.8997
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.18685 0.9243 0.8380
## Specificity 0.85066 0.9896 0.9641
## Pos Pred Value 0.17116 0.9398 0.8487
## Neg Pred Value 0.86373 0.9868 0.9612
## Prevalence 0.14167 0.1490 0.1936
## Detection Rate 0.02647 0.1377 0.1623
## Detection Prevalence 0.15466 0.1466 0.1912
## Balanced Accuracy 0.51875 0.9570 0.9011
db_tda_pc_5.50.5_n1_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 294 0 183 0 253 17 15
## BOMBAY 0 0 0 0 0 0 0
## CALI 36 78 29 0 209 9 9
## DERMASON 0 0 0 873 1 6 59
## HOROZ 60 78 274 72 108 0 39
## SEKER 0 0 0 30 0 562 6
## SIRA 6 0 3 88 7 14 662
##
## Overall Statistics
##
## Accuracy : 0.6196
## 95% CI : (0.6045, 0.6345)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.5418
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.74242 0.00000 0.059305 0.8213
## Specificity 0.87296 1.00000 0.905040 0.9781
## Pos Pred Value 0.38583 NaN 0.078378 0.9297
## Neg Pred Value 0.96926 0.96176 0.876011 0.9395
## Prevalence 0.09706 0.03824 0.119853 0.2605
## Detection Rate 0.07206 0.00000 0.007108 0.2140
## Detection Prevalence 0.18676 0.00000 0.090686 0.2301
## Balanced Accuracy 0.80769 0.50000 0.482173 0.8997
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.18685 0.9243 0.8380
## Specificity 0.85066 0.9896 0.9641
## Pos Pred Value 0.17116 0.9398 0.8487
## Neg Pred Value 0.86373 0.9868 0.9612
## Prevalence 0.14167 0.1490 0.1936
## Detection Rate 0.02647 0.1377 0.1623
## Detection Prevalence 0.15466 0.1466 0.1912
## Balanced Accuracy 0.51875 0.9570 0.9011
db_tda_pc_5.50.5_n1_db_nb_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.6196078 0.5418470 0.6045071 0.6345371 0.2605392
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
db_tda_pc_5.50.5_n1_db_nb_cf0_ov_acc<-db_tda_pc_5.50.5_n1_db_nb_cf0$overall[1]
db_tda_pc_5.50.5_n1_db_nb_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value
## Class: BARBUNYA 0.7424242 0.8729642 0.38582677 0.9692586
## Class: BOMBAY 0.0000000 1.0000000 NaN 0.9617647
## Class: CALI 0.0593047 0.9050404 0.07837838 0.8760108
## Class: DERMASON 0.8212606 0.9781240 0.92971246 0.9395097
## Class: HOROZ 0.1868512 0.8506568 0.17115689 0.8637286
## Class: SEKER 0.9243421 0.9896313 0.93979933 0.9867892
## Class: SIRA 0.8379747 0.9641337 0.84871795 0.9612121
## Precision Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.38582677 0.7424242 0.50777202 0.09705882 0.072058824
## Class: BOMBAY NA 0.0000000 NA 0.03823529 0.000000000
## Class: CALI 0.07837838 0.0593047 0.06752037 0.11985294 0.007107843
## Class: DERMASON 0.92971246 0.8212606 0.87212787 0.26053922 0.213970588
## Class: HOROZ 0.17115689 0.1868512 0.17866005 0.14166667 0.026470588
## Class: SEKER 0.93979933 0.9243421 0.93200663 0.14901961 0.137745098
## Class: SIRA 0.84871795 0.8379747 0.84331210 0.19362745 0.162254902
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.18676471 0.8076942
## Class: BOMBAY 0.00000000 0.5000000
## Class: CALI 0.09068627 0.4821725
## Class: DERMASON 0.23014706 0.8996923
## Class: HOROZ 0.15465686 0.5187540
## Class: SEKER 0.14656863 0.9569867
## Class: SIRA 0.19117647 0.9010542
db_tda_pc_5.50.5_n1_db_nb_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n1_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nb_n1_3_fold<-(db_nb_fit_re - db_tda_pc_5.50.5_n1_nb_fit_re)
diff_drybean_tda_pca_5.50.5_nb_n1_3_fold
## Accuracy
## 1 0.04418707
## 2 0.04525245
## 3 0.04067882
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0.75
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nb.n1_3_fold_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nb.n1_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nb.n1_3_fold
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.009466667
##
## $winRight
## [1] 0.9905333
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nb.n1_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nb.n1_3_fold
## $left
## [1] 0.0004461477
##
## $rope
## [1] 0.0006926052
##
## $right
## [1] 0.9988612
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold))
#bf_tda_pca_5.50.5_nb.n1_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nb_n1_3_fold)
## t = 31.392, df = 2, p-value = 0.001013
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.03742805 0.04931750
## sample estimates:
## mean of x
## 0.04337278
### Test set diff
diff_drybean_tda_pca_5.50.5_nb.n1_test<-(db_nb_cf_ov_acc - db_tda_pc_5.50.5_n1_db_nb_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nb.n1_test
## Accuracy
## 0.2747549
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n1_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nb.n1_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n1_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n1_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nb.n1_test$probRight
bst_dbf_db_tda_pca_5.50.5_nb.n1_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nb.n1_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n1_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nb.n1_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1603333
##
## $winRight
## [1] 0.8396667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nb.n1_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n1_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nb.n1_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb.n1_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb.n1_test)) #bf_tda_pca_5.50.5_nb.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n1_test))
##Node2
#DryBean_TDA_PC_5.50.5_n2_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_PC_5.50.5_n2_NbFit0
#DryBean_TDA_PC_5.50.5_n2_NbFit0$resample
#db_tda_pc_5.50.5_n2_nb_fit_re<-DryBean_TDA_PC_5.50.5_n2_NbFit0$resample[1]
#summary(DryBean_TDA_PC_5.50.5_n2_NbFit0)
# Predict outcome using DryBean_TDA_PC_5.50.5_n2_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_PC_5.50.5_n2_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#db_tda_pc_5.50.5_n2_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#db_tda_pc_5.50.5_n2_db_nb_cf0
#db_tda_pc_5.50.5_n2_db_nb_cf0
#db_tda_pc_5.50.5_n2_db_nb_cf0$overall
#db_tda_pc_5.50.5_n2_db_nb_cf0_ov_acc<-db_tda_pc_5.50.5_n2_db_nb_cf0$overall[1]
#db_tda_pc_5.50.5_n2_db_nb_cf0$byClass
#db_tda_pc_5.50.5_n2_db_nb_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n2_db_nb_cf0$byClass[5:7]#
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_pca_5.50.5_nb_n2_3_fold<-(db_nb_fit_re - db_tda_pc_5.50.5_n2_nb_fit_re)
#diff_drybean_tda_pca_5.50.5_nb_n2_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n2_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n2_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_nb.n2_3_fold
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n2_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n2_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold))
#bf_tda_pca_5.50.5_nb.n2_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n2_3_fold))
### Test set diff
#diff_drybean_tda_pca_5.50.5_nb.n2_test<-(db_nb_cf_ov_acc - db_tda_pc_5.50.5_n2_db_nb_cf0_ov_acc)
#diff_drybean_tda_pca_5.50.5_nb.n2_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n2_test<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n2_test
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n2_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n2_test$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n2_test$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n2_test_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n2_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_nb.n2_test
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n2_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n2_test
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb.n2_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test)) #bf_tda_pca_5.50.5_nb.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test))
##Node3
#DryBean_TDA_PC_5.50.5_n3_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_PC_5.50.5_n3_NbFit0
#DryBean_TDA_PC_5.50.5_n3_NbFit0$resample
#db_tda_pc_5.50.5_n3_nb_fit_re<-DryBean_TDA_PC_5.50.5_n3_NbFit0$resample[1]
#summary(DryBean_TDA_PC_5.50.5_n3_NbFit0)
# Predict outcome using DryBean_TDA_PC_5.50.5_n3_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_PC_5.50.5_n3_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#db_tda_pc_5.50.5_n3_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#db_tda_pc_5.50.5_n3_db_nb_cf0
#db_tda_pc_5.50.5_n3_db_nb_cf0
#db_tda_pc_5.50.5_n3_db_nb_cf0$overall
#db_tda_pc_5.50.5_n3_db_nb_cf0_ov_acc<-db_tda_pc_5.50.5_n3_db_nb_cf0$overall[1]
#db_tda_pc_5.50.5_n3_db_nb_cf0$byClass
#db_tda_pc_5.50.5_n3_db_nb_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n3_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_pca_5.50.5_nb_n3_3_fold<-(db_nb_fit_re - db_tda_pc_5.50.5_n3_nb_fit_re)
#diff_drybean_tda_pca_5.50.5_nb_n3_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n3_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n3_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_nb.n3_3_fold
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n3_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n3_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold))
#bf_tda_pca_5.50.5_nb.n3_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n3_3_fold))
### Test set diff
#diff_drybean_tda_pca_5.50.5_nb.n3_test<-(db_nb_cf_ov_acc - db_tda_pc_5.50.5_n3_db_nb_cf0_ov_acc)
#diff_drybean_tda_pca_5.50.5_nb.n3_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n3_test<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n3_test),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n3_test
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n3_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n3_test$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n3_test$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n3_test_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n2_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n2_test),-0.01,0.01)
##bsr_dbf_db_tda_pca_5.50.5_nb.n2_test
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n3_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n3_test),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n3_test
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb.n3_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb.n3_test)) #bf_tda_pca_5.50.5_nb.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n3_test))
##Node4
DryBean_TDA_PC_5.50.5_n4_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
method = 'nb',
trControl = fitControl,
metric='Accuracy')
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 23
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 48
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 57
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 68
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 94
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 48
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 54
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 57
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 68
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 94
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 83
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 83
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 61
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 86
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 99
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 61
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 86
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 99
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 525
DryBean_TDA_PC_5.50.5_n4_NbFit0
## Naive Bayes
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1061, 1060, 1059
## Resampling results across tuning parameters:
##
## usekernel Accuracy Kappa
## FALSE 0.9578615 0.9391147
## TRUE 0.9566013 0.9371819
##
## Tuning parameter 'fL' was held constant at a value of 0
## Tuning
## parameter 'adjust' was held constant at a value of 1
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were fL = 0, usekernel = FALSE and adjust
## = 1.
DryBean_TDA_PC_5.50.5_n4_NbFit0$resample
## Accuracy Kappa Resample
## 1 0.9565217 0.9373478 Fold1
## 2 0.9603774 0.9428498 Fold2
## 3 0.9566855 0.9371465 Fold3
db_tda_pc_5.50.5_n4_nb_fit_re<-DryBean_TDA_PC_5.50.5_n4_NbFit0$resample[1]
summary(DryBean_TDA_PC_5.50.5_n4_NbFit0)
## Length Class Mode
## apriori 4 table numeric
## tables 16 -none- list
## levels 4 -none- character
## call 5 -none- call
## x 16 data.frame list
## usekernel 1 -none- logical
## varnames 16 -none- character
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 3 data.frame list
## obsLevels 4 -none- character
## param 0 -none- list
# Predict outcome using DryBean_TDA_PC_5.50.5_n4_NbFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_PC_5.50.5_n4_NbFit0, newdata= Dry_Bean_DatasetTest)
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 5
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 6
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 7
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 8
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 9
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 10
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 11
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 12
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 13
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 14
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 15
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 16
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 17
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 18
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 19
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 20
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 21
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 22
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 23
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 24
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 25
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 26
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 27
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 28
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 29
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 30
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 31
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 32
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 33
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 34
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 35
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 37
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 39
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 48
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 51
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 53
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 54
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 55
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 56
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 57
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 58
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 59
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 60
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 61
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 62
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 63
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 64
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 65
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 67
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 68
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 69
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 70
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 71
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 72
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 73
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 74
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 75
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 76
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 77
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 78
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 79
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 80
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 81
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 82
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 83
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 84
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 85
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 86
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 87
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 88
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 89
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 90
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 91
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 92
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 93
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 94
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 95
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 96
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 97
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 98
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 99
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3080
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3082
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3084
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3086
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3087
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3088
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3090
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3091
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3093
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3094
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3095
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3096
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3097
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3098
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4080
# Create confusion matrix to assess model fit/performance on test data
db_tda_pc_5.50.5_n4_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
db_tda_pc_5.50.5_n4_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 81 330 1 600 314
## BOMBAY 1 156 0 0 0 8 0
## CALI 18 0 386 0 8 0 1
## DERMASON 0 0 0 0 0 0 0
## HOROZ 13 0 22 733 569 0 475
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3615
## 95% CI : (0.3468, 0.3765)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2771
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.78937 0.0000
## Specificity 0.64007 0.99771 0.99248 1.0000
## Pos Pred Value 0.21538 0.94545 0.93462 NaN
## Neg Pred Value 0.98661 1.00000 0.97191 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.08922 0.03824 0.09461 0.0000
## Detection Prevalence 0.41422 0.04044 0.10123 0.0000
## Balanced Accuracy 0.77963 0.99885 0.89092 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9844 0.000 0.0000
## Specificity 0.6451 1.000 1.0000
## Pos Pred Value 0.3140 NaN NaN
## Neg Pred Value 0.9960 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1395 0.000 0.0000
## Detection Prevalence 0.4441 0.000 0.0000
## Balanced Accuracy 0.8147 0.500 0.5000
db_tda_pc_5.50.5_n4_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 81 330 1 600 314
## BOMBAY 1 156 0 0 0 8 0
## CALI 18 0 386 0 8 0 1
## DERMASON 0 0 0 0 0 0 0
## HOROZ 13 0 22 733 569 0 475
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3615
## 95% CI : (0.3468, 0.3765)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2771
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.78937 0.0000
## Specificity 0.64007 0.99771 0.99248 1.0000
## Pos Pred Value 0.21538 0.94545 0.93462 NaN
## Neg Pred Value 0.98661 1.00000 0.97191 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.08922 0.03824 0.09461 0.0000
## Detection Prevalence 0.41422 0.04044 0.10123 0.0000
## Balanced Accuracy 0.77963 0.99885 0.89092 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9844 0.000 0.0000
## Specificity 0.6451 1.000 1.0000
## Pos Pred Value 0.3140 NaN NaN
## Neg Pred Value 0.9960 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1395 0.000 0.0000
## Detection Prevalence 0.4441 0.000 0.0000
## Balanced Accuracy 0.8147 0.500 0.5000
db_tda_pc_5.50.5_n4_db_nb_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.615196e-01 2.770842e-01 3.467586e-01 3.764792e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 6.574974e-46 NaN
db_tda_pc_5.50.5_n4_db_nb_cf0_ov_acc<-db_tda_pc_5.50.5_n4_db_nb_cf0$overall[1]
db_tda_pc_5.50.5_n4_db_nb_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.6400651 0.2153846 0.9866109 0.2153846
## Class: BOMBAY 1.0000000 0.9977064 0.9454545 1.0000000 0.9454545
## Class: CALI 0.7893661 0.9924812 0.9346247 0.9719116 0.9346247
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9844291 0.6450600 0.3140177 0.9960317 0.3140177
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.3489933 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 0.9719626 0.03823529 0.03823529
## Class: CALI 0.7893661 0.8558758 0.11985294 0.09460784
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9844291 0.4761506 0.14166667 0.13946078
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.41421569 0.7796285
## Class: BOMBAY 0.04044118 0.9988532
## Class: CALI 0.10122549 0.8909236
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.44411765 0.8147445
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
db_tda_pc_5.50.5_n4_db_nb_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n4_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_pca_5.50.5_nb_n4_3_fold<-(db_nb_fit_re - db_tda_pc_5.50.5_n4_nb_fit_re)
diff_drybean_tda_pca_5.50.5_nb_n4_3_fold
## Accuracy
## 1 -0.05973130
## 2 -0.05858321
## 3 -0.04956963
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold$probLeft/bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold$probRight
bst_dbf_db_tda_pca_5.50.5_nb.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nb.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nb.n4_3_fold
## $winLeft
## [1] 0.9901667
##
## $winRope
## [1] 0.009833333
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nb.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nb.n4_3_fold
## $left
## [1] 0.9967735
##
## $rope
## [1] 0.001652132
##
## $right
## [1] 0.001574345
# Rope Plot
plot(rope(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold))
#bf_tda_pca_5.50.5_nb.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold)
## t = -17.417, df = 2, p-value = 0.00328
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.06978587 -0.04213689
## sample estimates:
## mean of x
## -0.05596138
### Test set diff
diff_drybean_tda_pca_5.50.5_nb.n4_test<-(db_nb_cf_ov_acc - db_tda_pc_5.50.5_n4_db_nb_cf0_ov_acc)
diff_drybean_tda_pca_5.50.5_nb.n4_test
## Accuracy
## 0.5328431
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n4_test),-0.01,0.01)
bst_dbf_db_tda_pca_5.50.5_nb.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_dbf_db_tda_pca_5.50.5_nb.n4_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n4_test$probLeft/bst_dbf_db_tda_pca_5.50.5_nb.n4_test$probRight
bst_dbf_db_tda_pca_5.50.5_nb.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_dbf_db_tda_pca_5.50.5_nb.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n4_test),-0.01,0.01)
bsr_dbf_db_tda_pca_5.50.5_nb.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.1589333
##
## $winRight
## [1] 0.8410667
# Bayesian Correlated Test
bct_dbf_db_tda_pca_5.50.5_nb.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n4_test),0.1,-0.01,0.01)
bct_dbf_db_tda_pca_5.50.5_nb.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb.n4_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb.n4_test)) #bf_tda_pca_5.50.5_nb.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n4_test))
##Node5
#DryBean_TDA_PC_5.50.5_n5_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_PC_5.50.5_n5_NbFit0
#DryBean_TDA_PC_5.50.5_n5_NbFit0$resample
#db_tda_pc_5.50.5_n5_nb_fit_re<-DryBean_TDA_PC_5.50.5_n5_NbFit0$resample[1]
#summary(DryBean_TDA_PC_5.50.5_n5_NbFit0)
# Predict outcome using DryBean_TDA_PC_5.50.5_n5_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_PC_5.50.5_n5_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#db_tda_pc_5.50.5_n5_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#db_tda_pc_5.50.5_n5_db_nb_cf0
#db_tda_pc_5.50.5_n5_db_nb_cf0
#db_tda_pc_5.50.5_n5_db_nb_cf0$overall
#db_tda_pc_5.50.5_n5_db_nb_cf0_ov_acc<-db_tda_pc_5.50.5_n5_db_nb_cf0$overall[1]
#db_tda_pc_5.50.5_n5_db_nb_cf0$byClass
#db_tda_pc_5.50.5_n5_db_nb_cf0_pre_rec_f1<-db_tda_pc_5.50.5_n5_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_pca_5.50.5_nb_n5_3_fold<-(db_nb_fit_re - db_tda_pc_5.50.5_n5_nb_fit_re)
#diff_drybean_tda_pca_5.50.5_nb_n5_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n5_3_fold),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n5_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n5_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n4_3_fold),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_nb.n5_3_fold
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n5_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n5_3_fold),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n5_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb_n5_3_fold))
#bf_tda_pca_5.50.5_nb.n5_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb_n5_3_fold))
### Test set diff
#diff_drybean_tda_pca_5.50.5_nb.n5_test<-(db_nb_cf_ov_acc - db_tda_pc_5.50.5_n5_db_nb_cf0_ov_acc)
#diff_drybean_tda_pca_5.50.5_nb.n5_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n5_test<-#BayesianSignTest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n5_test),-0.01,0.01)
#bst_dbf_db_tda_pca_5.50.5_nb.n5_test
# Odds Left Bayesian Sign Test
#bst_dbf_db_tda_pca_5.50.5_nb.n5_test_odds.left<-bst_dbf_db_tda_pca_5.50.5_nb.n5_test$probLeft/#bst_dbf_db_tda_pca_5.50.5_nb.n5_test$probRight
#bst_dbf_db_tda_pca_5.50.5_nb.n5_test_odds.left
# Bayesian Signed Rank Test
#bsr_dbf_db_tda_pca_5.50.5_nb.n5_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n5_test),-0.01,0.01)
#bsr_dbf_db_tda_pca_5.50.5_nb.n5_test
# Bayesian Correlated Test
#bct_dbf_db_tda_pca_5.50.5_nb.n5_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n5_test),0.1,-0.01,0.01)
#bct_dbf_db_tda_pca_5.50.5_nb.n5_test
# Rope Plot
#plot(rope(diff_drybean_tda_pca_5.50.5_nb.n5_test)))
#BayesFactor
#bf_tda_pca_5.50.5_nb.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_pca_5.50.5_nb.n5_test)) #bf_tda_pca_5.50.5_nb.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_pca_5.50.5_nb.n5_test))
##With TDA KDE filter 5 intervals, 50% overlap, 5 bins
##Node1
#DryBean_TDA_KDE_5.50.5_n1_NbFit0 <- train(as.factor(Class) ~ ., data = #tda.m_kde_dry_bean_dataset_5.50.5.n5.n1.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_KDE_5.50.5_n1_NbFit0
#DryBean_TDA_KDE_5.50.5_n1_NbFit0$resample
#nb_tda_kde_5.50.5_n1_nb_fit_re<-DryBean_TDA_KDE_5.50.5_n1_NbFit0$resample[1]
#summary(DryBean_TDA_KDE_5.50.5_n1_NbFit0)
# Predict outcome using DryBean_TDA_KDE_5.50.5_n1_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_KDE_5.50.5_n1_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#nb_tda_kde_5.50.5_n1_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#nb_tda_kde_5.50.5_n1_db_nb_cf0
#nb_tda_kde_5.50.5_n1_db_nb_cf0
#nb_tda_kde_5.50.5_n1_db_nb_cf0$overall
#nb_tda_kde_5.50.5_n1_db_nb_cf0_ov_acc<-nb_tda_kde_5.50.5_n1_db_nb_cf0$overall[1]
#nb_tda_kde_5.50.5_n1_db_nb_cf0$byClas1
#nb_tda_kde_5.50.5_n1_db_nb_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n1_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_kde_5.50.5_nb_n1_3_fold<-(db_nb_fit_re - nb_tda_kde_5.50.5_n1_nb_fit_re)
#diff_drybean_tda_kde_5.50.5_nb_n1_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n1_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n1_3_fold
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n1_3_fold_odds.left<-bst_tda_kde_5.50.5_nb.n1_3_fold$probLeft/#bst_tda_kde_5.50.5_nb.n1_3_fold$probRight
#bst_tda_kde_5.50.5_nb.n1_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n1_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n1_3_fold
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nb.n1_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n1_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n1_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold))
#bf_tda_kde_5.50.5_nb.n1_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n1_3_fold))
### Test set diff
#diff_drybean_tda_kde_5.50.5_nb.n1_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n1_db_nb_cf0_ov_acc)
#diff_drybean_tda_kde_5.50.5_nb.n1_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n1_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n1_test),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n1_test
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n1_test_odds.left<-bst_tda_kde_5.50.5_nb.n1_test$probLeft/#bst_tda_kde_5.50.5_nb.n1_test$probRight
#bst_tda_kde_5.50.5_nb.n1_test_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n1_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n1_test),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n1_test
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n1_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n1_test),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n1_test
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb.n1_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n1_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb.n1_test)) #bf_tda_pca_5.50.5_nb.n1_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n1_test))
##Node2
#DryBean_TDA_KDE_5.50.5_n2_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n2.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_KDE_5.50.5_n2_NbFit0
#DryBean_TDA_KDE_5.50.5_n2_NbFit0$resample
#nb_tda_kde_5.50.5_n2_nb_fit_re<-DryBean_TDA_KDE_5.50.5_n2_NbFit0$resample[1]
#summary(DryBean_TDA_KDE_5.50.5_n2_NbFit0)
# Predict outcome using DryBean_TDA_KDE_5.50.5_n2_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_KDE_5.50.5_n2_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#nb_tda_kde_5.50.5_n2_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#nb_tda_kde_5.50.5_n2_db_nb_cf0
#nb_tda_kde_5.50.5_n2_db_nb_cf0
#nb_tda_kde_5.50.5_n2_db_nb_cf0$overall
#nb_tda_kde_5.50.5_n2_db_nb_cf0_ov_acc<-nb_tda_kde_5.50.5_n2_db_nb_cf0$overall[1]
#nb_tda_kde_5.50.5_n2_db_nb_cf0$byClass
#nb_tda_kde_5.50.5_n2_db_nb_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n2_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_kde_5.50.5_nb_n2_3_fold<-(db_nb_fit_re - nb_tda_kde_5.50.5_n2_nb_fit_re)
#diff_drybean_tda_kde_5.50.5_nb_n2_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n2_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n2_3_fold
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n2_3_fold_odds.left<-bst_tda_kde_5.50.5_nb.n2_3_fold$probLeft/#bst_tda_kde_5.50.5_nb.n2_3_fold$probRight
#bst_tda_kde_5.50.5_nb.n2_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n2_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n2_3_fold
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n2_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n2_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n2_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold))
#bf_tda_kde_5.50.5_nb.n2_3_fold
#t_test
t#.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n2_3_fold))
### Test set diff
#diff_drybean_tda_kde_5.50.5_nb.n2_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n2_db_nb_cf0_ov_acc)
#diff_drybean_tda_kde_5.50.5_nb.n2_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n2_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n2_test),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n2_test
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n2_test_odds.left<-bst_tda_kde_5.50.5_nb.n2_test$probLeft/#bst_tda_kde_5.50.5_nb.n2_test$probRight
#bst_tda_kde_5.50.5_nb.n2_test_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n2_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n2_test),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n2_test
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n2_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n2_test),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n2_test
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb.n2_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n2_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb.n2_test)) #bf_tda_kde_5.50.5_nb.n2_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n2_test))
##Node3
#DryBean_TDA_KDE_5.50.5_n3_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n3.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_KDE_5.50.5_n3_NbFit0
#DryBean_TDA_KDE_5.50.5_n3_NbFit0$resample
#nb_tda_kde_5.50.5_n3_nb_fit_re<-DryBean_TDA_KDE_5.50.5_n3_NbFit0$resample[1]
#summary(DryBean_TDA_KDE_5.50.5_n3_NbFit0)
# Predict outcome using DryBean_TDA_KDE_5.50.5_n3_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_KDE_5.50.5_n3_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#nb_tda_kde_5.50.5_n3_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#nb_tda_kde_5.50.5_n3_db_nb_cf0
##nb_tda_kde_5.50.5_n3_db_nb_cf0
#nb_tda_kde_5.50.5_n3_db_nb_cf0$overall
#nb_tda_kde_5.50.5_n3_db_nb_cf0_ov_acc<-nb_tda_kde_5.50.5_n3_db_nb_cf0$overall[1]
#nb_tda_kde_5.50.5_n3_db_nb_cf0$byClass
#nb_tda_kde_5.50.5_n3_db_nb_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n3_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_kde_5.50.5_nb_n3_3_fold<-(db_nb_fit_re - nb_tda_kde_5.50.5_n3_nb_fit_re)
#diff_drybean_tda_kde_5.50.5_nb_n3_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n3_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n3_3_fold
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n3_3_fold_odds.left<-bst_tda_kde_5.50.5_nb.n3_3_fold$probLeft/#bst_tda_kde_5.50.5_nb.n3_3_fold$probRight
#bst_tda_kde_5.50.5_nb.n3_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n3_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n3_3_fold
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n3_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n3_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n3_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold))
#bf_tda_kde_5.50.5_nb.n3_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n3_3_fold))
### Test set diff
#diff_drybean_tda_kde_5.50.5_nb.n3_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n3_db_nb_cf0_ov_acc)
#diff_drybean_tda_kde_5.50.5_nb.n3_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n3_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n3_test),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n3_test
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n3_test_odds.left<-bst_tda_kde_5.50.5_nb.n3_test$probLeft/#bst_tda_kde_5.50.5_nb.n3_test$probRight
#bst_tda_kde_5.50.5_nb.n3_test_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n3_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n3_test),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n3_test
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n3_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n3_test),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n3_test
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb.n3_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n3_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb.n3_test)) #bf_tda_kde_5.50.5_nb.n3_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n3_test))
##Node4
DryBean_TDA_KDE_5.50.5_n4_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n4.vec,
method = 'nb',
trControl = fitControl,
metric='Accuracy')
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 9
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 33
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 35
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 60
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 77
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 90
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 97
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 33
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 35
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 60
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 77
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 90
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 97
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 95
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 54
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 95
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 23
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 51
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 73
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 51
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 54
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 73
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 441
DryBean_TDA_KDE_5.50.5_n4_NbFit0
## Naive Bayes
##
## 1590 samples
## 16 predictor
## 4 classes: 'BARBUNYA', 'BOMBAY', 'CALI', 'HOROZ'
##
## No pre-processing
## Resampling: Cross-Validated (3 fold)
## Summary of sample sizes: 1060, 1059, 1061
## Resampling results across tuning parameters:
##
## usekernel Accuracy Kappa
## FALSE 0.9565835 0.9373926
## TRUE 0.9540749 0.9334847
##
## Tuning parameter 'fL' was held constant at a value of 0
## Tuning
## parameter 'adjust' was held constant at a value of 1
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were fL = 0, usekernel = FALSE and adjust
## = 1.
DryBean_TDA_KDE_5.50.5_n4_NbFit0$resample
## Accuracy Kappa Resample
## 1 0.9622642 0.9452660 Fold1
## 2 0.9698682 0.9563098 Fold2
## 3 0.9376181 0.9106021 Fold3
nb_tda_kde_5.50.5_n4_nb_fit_re<-DryBean_TDA_KDE_5.50.5_n4_NbFit0$resample[1]
summary(DryBean_TDA_KDE_5.50.5_n4_NbFit0)
## Length Class Mode
## apriori 4 table numeric
## tables 16 -none- list
## levels 4 -none- character
## call 5 -none- call
## x 16 data.frame list
## usekernel 1 -none- logical
## varnames 16 -none- character
## xNames 16 -none- character
## problemType 1 -none- character
## tuneValue 3 data.frame list
## obsLevels 4 -none- character
## param 0 -none- list
# Predict outcome using DryBean_TDA_KDE_5.50.5_n4_NbFit0 from training data based on testing data
pred0 <- predict(DryBean_TDA_KDE_5.50.5_n4_NbFit0, newdata= Dry_Bean_DatasetTest)
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 5
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 6
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 7
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 8
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 9
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 10
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 11
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 12
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 13
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 14
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 15
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 16
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 17
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 18
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 19
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 20
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 21
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 22
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 23
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 24
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 25
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 26
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 27
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 28
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 29
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 30
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 31
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 32
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 33
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 34
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 35
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 36
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 37
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 38
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 39
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 40
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 41
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 42
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 43
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 44
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 45
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 46
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 47
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 48
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 49
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 50
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 51
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 52
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 53
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 54
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 55
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 56
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 57
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 58
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 59
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 60
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 61
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 62
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 63
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 64
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 65
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 66
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 67
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 68
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 69
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 70
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 71
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 72
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 73
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 74
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 75
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 76
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 77
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 78
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 79
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 80
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 81
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 82
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 83
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 84
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 85
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 86
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 87
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 88
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 89
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 90
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 91
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 92
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 93
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 94
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 95
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 96
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 97
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 98
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 99
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 1971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 2999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3080
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3081
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3082
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3083
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3084
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3085
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3086
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3087
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3088
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3089
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3090
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3091
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3092
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3093
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3094
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3095
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3096
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3097
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3098
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3099
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3100
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3101
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3102
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3103
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3104
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3105
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3106
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3107
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3108
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3109
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3110
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3111
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3112
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3113
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3114
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3115
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3116
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3117
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3118
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3119
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3120
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3121
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3122
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3123
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3124
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3125
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3126
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3127
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3128
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3129
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3130
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3131
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3132
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3133
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3134
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3135
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3136
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3137
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3138
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3139
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3140
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3141
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3142
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3143
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3144
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3145
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3146
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3147
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3148
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3149
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3150
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3151
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3152
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3153
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3154
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3155
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3156
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3157
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3158
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3159
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3160
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3161
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3162
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3163
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3164
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3165
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3166
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3167
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3168
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3169
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3170
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3171
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3172
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3173
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3174
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3175
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3176
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3177
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3178
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3179
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3180
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3181
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3182
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3183
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3184
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3185
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3186
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3187
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3188
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3189
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3190
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3191
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3192
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3193
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3194
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3195
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3196
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3197
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3198
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3199
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3200
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3201
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3202
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3203
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3204
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3205
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3206
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3207
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3208
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3209
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3210
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3211
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3212
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3213
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3214
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3215
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3216
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3217
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3218
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3219
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3220
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3221
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3222
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3223
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3224
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3225
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3226
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3227
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3228
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3229
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3230
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3231
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3232
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3233
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3234
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3235
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3236
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3237
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3238
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3239
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3240
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3241
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3242
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3243
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3244
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3245
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3246
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3247
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3248
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3249
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3250
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3251
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3252
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3253
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3254
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3255
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3256
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3257
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3258
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3259
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3260
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3261
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3262
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3263
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3264
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3265
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3266
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3267
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3268
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3269
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3270
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3271
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3272
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3273
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3274
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3275
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3276
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3277
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3278
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3279
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3280
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3281
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3282
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3283
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3284
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3285
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3286
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3287
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3288
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3289
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3290
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3291
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3292
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3293
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3294
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3295
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3296
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3297
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3298
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3299
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3300
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3301
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3302
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3303
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3304
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3305
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3306
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3307
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3308
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3309
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3310
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3311
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3312
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3313
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3314
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3315
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3316
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3317
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3318
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3319
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3320
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3321
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3322
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3323
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3324
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3325
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3326
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3327
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3328
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3329
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3330
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3331
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3332
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3333
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3334
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3335
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3336
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3337
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3338
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3339
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3340
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3341
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3342
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3343
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3344
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3345
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3346
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3347
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3348
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3349
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3350
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3351
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3352
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3353
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3354
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3355
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3356
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3357
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3358
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3359
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3360
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3361
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3362
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3363
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3364
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3365
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3366
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3367
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3368
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3369
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3370
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3371
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3372
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3373
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3374
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3375
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3376
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3377
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3378
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3379
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3380
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3381
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3382
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3383
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3384
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3385
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3386
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3387
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3388
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3389
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3390
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3391
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3392
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3393
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3394
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3395
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3396
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3397
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3398
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3399
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3400
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3401
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3402
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3403
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3404
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3405
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3406
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3407
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3408
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3409
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3410
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3411
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3412
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3413
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3414
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3415
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3416
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3417
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3418
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3419
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3420
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3421
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3422
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3423
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3424
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3425
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3426
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3427
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3428
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3429
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3430
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3431
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3432
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3433
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3434
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3435
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3436
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3437
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3438
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3439
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3440
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3441
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3442
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3443
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3444
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3445
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3446
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3447
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3448
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3449
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3450
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3451
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3452
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3453
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3454
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3455
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3456
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3457
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3458
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3459
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3460
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3461
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3462
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3463
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3464
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3465
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3466
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3467
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3468
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3469
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3470
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3471
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3472
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3473
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3474
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3475
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3476
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3477
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3478
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3479
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3480
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3481
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3482
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3483
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3484
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3485
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3486
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3487
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3488
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3489
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3490
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3491
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3492
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3493
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3494
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3495
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3496
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3497
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3498
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3499
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3500
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3501
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3502
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3503
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3504
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3505
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3506
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3507
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3508
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3509
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3510
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3511
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3512
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3513
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3514
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3515
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3516
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3517
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3518
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3519
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3520
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3521
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3522
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3523
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3524
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3525
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3526
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3527
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3528
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3529
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3530
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3531
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3532
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3533
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3534
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3535
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3536
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3537
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3538
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3539
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3540
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3541
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3542
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3543
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3544
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3545
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3546
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3547
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3548
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3549
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3550
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3551
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3552
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3553
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3554
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3555
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3556
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3557
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3558
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3559
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3560
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3561
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3562
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3563
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3564
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3565
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3566
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3567
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3568
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3569
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3570
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3571
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3572
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3573
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3574
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3575
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3576
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3577
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3578
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3579
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3580
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3581
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3582
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3583
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3584
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3585
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3586
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3587
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3588
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3589
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3590
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3591
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3592
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3593
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3594
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3595
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3596
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3597
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3598
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3599
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3600
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3601
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3602
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3603
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3604
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3605
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3606
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3607
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3608
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3609
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3610
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3611
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3612
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3613
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3614
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3615
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3616
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3617
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3618
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3619
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3620
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3621
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3622
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3623
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3624
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3625
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3626
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3627
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3628
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3629
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3630
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3631
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3632
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3633
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3634
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3635
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3636
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3637
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3638
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3639
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3640
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3641
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3642
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3643
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3644
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3645
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3646
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3647
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3648
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3649
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3650
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3651
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3652
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3653
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3654
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3655
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3656
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3657
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3658
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3659
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3660
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3661
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3662
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3663
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3664
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3665
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3666
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3667
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3668
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3669
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3670
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3671
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3672
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3673
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3674
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3675
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3676
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3677
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3678
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3679
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3680
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3681
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3682
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3683
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3684
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3685
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3686
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3687
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3688
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3689
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3690
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3691
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3692
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3693
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3694
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3695
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3696
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3697
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3698
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3699
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3700
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3701
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3702
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3703
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3704
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3705
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3706
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3707
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3708
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3709
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3710
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3711
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3712
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3713
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3714
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3715
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3716
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3717
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3718
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3719
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3720
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3721
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3722
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3723
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3724
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3725
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3726
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3727
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3728
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3729
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3730
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3731
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3732
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3733
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3734
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3735
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3736
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3737
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3738
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3739
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3740
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3741
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3742
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3743
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3744
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3745
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3746
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3747
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3748
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3749
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3750
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3751
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3752
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3753
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3754
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3755
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3756
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3757
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3758
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3759
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3760
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3761
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3762
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3763
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3764
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3765
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3766
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3767
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3768
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3769
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3770
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3771
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3772
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3773
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3774
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3775
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3776
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3777
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3778
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3779
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3780
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3781
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3782
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3783
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3784
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3785
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3786
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3787
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3788
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3789
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3790
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3791
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3792
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3793
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3794
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3795
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3796
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3797
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3798
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3799
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3800
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3801
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3802
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3803
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3804
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3805
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3806
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3807
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3808
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3809
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3810
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3811
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3812
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3813
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3814
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3815
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3816
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3817
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3818
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3819
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3820
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3821
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3822
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3823
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3824
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3825
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3826
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3827
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3828
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3829
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3830
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3831
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3832
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3833
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3834
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3835
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3836
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3837
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3838
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3839
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3840
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3841
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3842
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3843
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3844
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3845
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3846
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3847
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3848
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3849
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3850
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3851
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3852
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3853
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3854
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3855
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3856
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3857
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3858
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3859
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3860
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3861
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3862
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3863
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3864
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3865
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3866
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3867
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3868
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3869
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3870
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3871
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3872
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3873
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3874
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3875
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3876
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3877
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3878
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3879
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3880
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3881
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3882
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3883
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3884
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3885
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3886
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3887
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3888
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3889
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3890
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3891
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3892
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3893
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3894
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3895
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3896
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3897
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3898
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3899
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3900
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3901
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3902
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3903
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3904
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3905
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3906
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3907
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3908
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3909
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3910
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3911
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3912
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3913
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3914
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3915
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3916
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3917
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3918
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3919
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3920
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3921
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3922
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3923
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3924
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3925
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3926
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3927
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3928
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3929
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3930
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3931
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3932
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3933
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3934
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3935
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3936
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3937
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3938
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3939
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3940
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3941
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3942
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3943
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3944
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3945
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3946
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3947
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3948
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3949
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3950
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3951
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3952
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3953
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3954
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3955
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3956
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3957
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3958
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3959
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3960
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3961
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3962
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3963
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3964
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3965
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3966
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3967
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3968
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3969
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3970
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3971
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3972
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3973
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3974
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3975
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3976
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3977
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3978
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3979
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3980
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3981
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3982
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3983
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3984
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3985
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3986
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3987
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3988
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3989
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3990
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3991
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3992
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3993
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3994
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3995
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3996
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3997
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3998
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 3999
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4000
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4001
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4002
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4003
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4004
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4005
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4006
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4007
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4008
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4009
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4010
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4011
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4012
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4013
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4014
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4015
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4016
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4017
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4018
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4019
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4020
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4021
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4022
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4023
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4024
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4025
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4026
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4027
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4028
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4029
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4030
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4031
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4032
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4033
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4034
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4035
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4036
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4037
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4038
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4039
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4040
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4041
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4042
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4043
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4044
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4045
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4046
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4047
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4048
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4049
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4050
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4051
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4052
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4053
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4054
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4055
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4056
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4057
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4058
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4059
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4060
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4061
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4062
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4063
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4064
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4065
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4066
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4067
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4068
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4069
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4070
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4071
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4072
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4073
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4074
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4075
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4076
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4077
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4078
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4079
## Warning in FUN(X[[i]], ...): Numerical 0 probability for all classes with
## observation 4080
# Create confusion matrix to assess model fit/performance on test data
nb_tda_kde_5.50.5_n4_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
## Warning in levels(reference) != levels(data): longer object length is not a
## multiple of shorter object length
## Warning in confusionMatrix.default(data = pred0,
## as.factor(Dry_Bean_DatasetTest$Class)): Levels are not in the same order for
## reference and data. Refactoring data to match.
nb_tda_kde_5.50.5_n4_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 81 330 1 600 314
## BOMBAY 1 156 0 0 0 8 0
## CALI 18 0 386 0 8 0 1
## DERMASON 0 0 0 0 0 0 0
## HOROZ 13 0 22 733 569 0 475
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3615
## 95% CI : (0.3468, 0.3765)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2771
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.78937 0.0000
## Specificity 0.64007 0.99771 0.99248 1.0000
## Pos Pred Value 0.21538 0.94545 0.93462 NaN
## Neg Pred Value 0.98661 1.00000 0.97191 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.08922 0.03824 0.09461 0.0000
## Detection Prevalence 0.41422 0.04044 0.10123 0.0000
## Balanced Accuracy 0.77963 0.99885 0.89092 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9844 0.000 0.0000
## Specificity 0.6451 1.000 1.0000
## Pos Pred Value 0.3140 NaN NaN
## Neg Pred Value 0.9960 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1395 0.000 0.0000
## Detection Prevalence 0.4441 0.000 0.0000
## Balanced Accuracy 0.8147 0.500 0.5000
nb_tda_kde_5.50.5_n4_db_nb_cf0
## Confusion Matrix and Statistics
##
## Reference
## Prediction BARBUNYA BOMBAY CALI DERMASON HOROZ SEKER SIRA
## BARBUNYA 364 0 81 330 1 600 314
## BOMBAY 1 156 0 0 0 8 0
## CALI 18 0 386 0 8 0 1
## DERMASON 0 0 0 0 0 0 0
## HOROZ 13 0 22 733 569 0 475
## SEKER 0 0 0 0 0 0 0
## SIRA 0 0 0 0 0 0 0
##
## Overall Statistics
##
## Accuracy : 0.3615
## 95% CI : (0.3468, 0.3765)
## No Information Rate : 0.2605
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.2771
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: BARBUNYA Class: BOMBAY Class: CALI Class: DERMASON
## Sensitivity 0.91919 1.00000 0.78937 0.0000
## Specificity 0.64007 0.99771 0.99248 1.0000
## Pos Pred Value 0.21538 0.94545 0.93462 NaN
## Neg Pred Value 0.98661 1.00000 0.97191 0.7395
## Prevalence 0.09706 0.03824 0.11985 0.2605
## Detection Rate 0.08922 0.03824 0.09461 0.0000
## Detection Prevalence 0.41422 0.04044 0.10123 0.0000
## Balanced Accuracy 0.77963 0.99885 0.89092 0.5000
## Class: HOROZ Class: SEKER Class: SIRA
## Sensitivity 0.9844 0.000 0.0000
## Specificity 0.6451 1.000 1.0000
## Pos Pred Value 0.3140 NaN NaN
## Neg Pred Value 0.9960 0.851 0.8064
## Prevalence 0.1417 0.149 0.1936
## Detection Rate 0.1395 0.000 0.0000
## Detection Prevalence 0.4441 0.000 0.0000
## Balanced Accuracy 0.8147 0.500 0.5000
nb_tda_kde_5.50.5_n4_db_nb_cf0$overall
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 3.615196e-01 2.770842e-01 3.467586e-01 3.764792e-01 2.605392e-01
## AccuracyPValue McnemarPValue
## 6.574974e-46 NaN
nb_tda_kde_5.50.5_n4_db_nb_cf0_ov_acc<-nb_tda_kde_5.50.5_n4_db_nb_cf0$overall[1]
nb_tda_kde_5.50.5_n4_db_nb_cf0$byClass
## Sensitivity Specificity Pos Pred Value Neg Pred Value Precision
## Class: BARBUNYA 0.9191919 0.6400651 0.2153846 0.9866109 0.2153846
## Class: BOMBAY 1.0000000 0.9977064 0.9454545 1.0000000 0.9454545
## Class: CALI 0.7893661 0.9924812 0.9346247 0.9719116 0.9346247
## Class: DERMASON 0.0000000 1.0000000 NaN 0.7394608 NA
## Class: HOROZ 0.9844291 0.6450600 0.3140177 0.9960317 0.3140177
## Class: SEKER 0.0000000 1.0000000 NaN 0.8509804 NA
## Class: SIRA 0.0000000 1.0000000 NaN 0.8063725 NA
## Recall F1 Prevalence Detection Rate
## Class: BARBUNYA 0.9191919 0.3489933 0.09705882 0.08921569
## Class: BOMBAY 1.0000000 0.9719626 0.03823529 0.03823529
## Class: CALI 0.7893661 0.8558758 0.11985294 0.09460784
## Class: DERMASON 0.0000000 NA 0.26053922 0.00000000
## Class: HOROZ 0.9844291 0.4761506 0.14166667 0.13946078
## Class: SEKER 0.0000000 NA 0.14901961 0.00000000
## Class: SIRA 0.0000000 NA 0.19362745 0.00000000
## Detection Prevalence Balanced Accuracy
## Class: BARBUNYA 0.41421569 0.7796285
## Class: BOMBAY 0.04044118 0.9988532
## Class: CALI 0.10122549 0.8909236
## Class: DERMASON 0.00000000 0.5000000
## Class: HOROZ 0.44411765 0.8147445
## Class: SEKER 0.00000000 0.5000000
## Class: SIRA 0.00000000 0.5000000
nb_tda_kde_5.50.5_n4_db_nb_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n4_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
diff_drybean_tda_kde_5.50.5_nb_n4_3_fold<-(db_nb_fit_re - nb_tda_kde_5.50.5_n4_nb_fit_re)
diff_drybean_tda_kde_5.50.5_nb_n4_3_fold
## Accuracy
## 1 -0.06547372
## 2 -0.06807403
## 3 -0.03050228
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nb.n4_3_fold<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold),-0.01,0.01)
bst_tda_kde_5.50.5_nb.n4_3_fold
## $probLeft
## [1] 0.75
##
## $probRope
## [1] 0.25
##
## $probRight
## [1] 0
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nb.n4_3_fold_odds.left<-bst_tda_kde_5.50.5_nb.n4_3_fold$probLeft/bst_tda_kde_5.50.5_nb.n4_3_fold$probRight
bst_tda_kde_5.50.5_nb.n4_3_fold_odds.left
## [1] Inf
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nb.n4_3_fold<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold),-0.01,0.01)
bsr_tda_kde_5.50.5_nb.n4_3_fold
## $winLeft
## [1] 0.9918
##
## $winRope
## [1] 0.0082
##
## $winRight
## [1] 0
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nb.n4_3_fold<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nb.n4_3_fold
## $left
## [1] 0.9571996
##
## $rope
## [1] 0.02094023
##
## $right
## [1] 0.02186021
# Rope Plot
plot(rope(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n4_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold))
#bf_tda_kde_5.50.5_nb.n4_3_fold
#t_test
t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold))
##
## One Sample t-test
##
## data: as.matrix(diff_drybean_tda_kde_5.50.5_nb_n4_3_fold)
## t = -4.5141, df = 2, p-value = 0.04573
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.106804863 -0.002561819
## sample estimates:
## mean of x
## -0.05468334
### Test set diff
diff_drybean_tda_kde_5.50.5_nb.n4_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n4_db_nb_cf0_ov_acc)
diff_drybean_tda_kde_5.50.5_nb.n4_test
## Accuracy
## 0.5654412
## Bayesian Tests Test set diff
# Bayesian Sign Test
bst_tda_kde_5.50.5_nb.n4_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n4_test),-0.01,0.01)
bst_tda_kde_5.50.5_nb.n4_test
## $probLeft
## [1] 0
##
## $probRope
## [1] 0.5
##
## $probRight
## [1] 0.5
# Odds Left Bayesian Sign Test
bst_tda_kde_5.50.5_nb.n4_test_odds.left<-bst_tda_kde_5.50.5_nb.n4_test$probLeft/bst_tda_kde_5.50.5_nb.n4_test$probRight
bst_tda_kde_5.50.5_nb.n4_test_odds.left
## [1] 0
# Bayesian Signed Rank Test
bsr_tda_kde_5.50.5_nb.n4_test<-BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n4_test),-0.01,0.01)
bsr_tda_kde_5.50.5_nb.n4_test
## $winLeft
## [1] 0
##
## $winRope
## [1] 0.159
##
## $winRight
## [1] 0.841
# Bayesian Correlated Test
bct_tda_kde_5.50.5_nb.n4_test<-correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n4_test),0.1,-0.01,0.01)
bct_tda_kde_5.50.5_nb.n4_test
## $left
## [1] NA
##
## $rope
## [1] NA
##
## $right
## [1] NA
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb.n4_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n4_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb.n4_test)) #bf_tda_kde_5.50.5_nb.n4_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n4_test))
##Node5
#DryBean_TDA_KDE_5.50.5_n5_NbFit0 <- train(as.factor(Class) ~ ., data = tda.m_dry_bean_dataset_5.50.5.n5.vec,
# method = 'nb',
# trControl = fitControl,
# metric='Accuracy')
#DryBean_TDA_KDE_5.50.5_n5_NbFit0
#DryBean_TDA_KDE_5.50.5_n5_NbFit0$resample
#nb_tda_kde_5.50.5_n5_nb_fit_re<-DryBean_TDA_KDE_5.50.5_n5_NbFit0$resample[1]
#summary(DryBean_TDA_KDE_5.50.5_n5_NbFit0)
# Predict outcome using DryBean_TDA_KDE_5.50.5_n5_NbFit0 from training data based on testing data
#pred0 <- predict(DryBean_TDA_KDE_5.50.5_n5_NbFit0, newdata= Dry_Bean_DatasetTest)
# Create confusion matrix to assess model fit/performance on test data
#nb_tda_kde_5.50.5_n5_db_nb_cf0<-confusionMatrix(data=pred0, as.factor(Dry_Bean_DatasetTest$Class))
#nb_tda_kde_5.50.5_n5_db_nb_cf0
#nb_tda_kde_5.50.5_n5_db_nb_cf0
#nb_tda_kde_5.50.5_n5_db_nb_cf0$overall
#nb_tda_kde_5.50.5_n5_db_nb_cf0_ov_acc<-nb_tda_kde_5.50.5_n5_db_nb_cf0$overall[1]
#nb_tda_kde_5.50.5_n5_db_nb_cf0$byClass
#nb_tda_kde_5.50.5_n5_db_nb_cf0_pre_rec_f1<-nb_tda_kde_5.50.5_n5_db_nb_cf0$byClass[5:7]
###### Conduct initial Bayesian tests of non-tda-assisted RF vs. tda-assisted RF classifiers
### 3-fold diff
#diff_drybean_tda_kde_5.50.5_nb_n5_3_fold<-(db_nb_fit_re - nb_tda_kde_5.50.5_n5_nb_fit_re)
#diff_drybean_tda_kde_5.50.5_nb_n5_3_fold
## Bayesian Tests 3-fold diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n5_3_fold<-#BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n5_3_fold
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n5_3_fold_odds.left<-bst_tda_kde_5.50.5_nb.n5_3_fold$probLeft/#bst_tda_kde_5.50.5_nb.n5_3_fold$probRight
#bst_tda_kde_5.50.5_nb.n5_3_fold_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n5_3_fold<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n5_3_fold
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n5_3_fold<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n5_3_fold
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold,c(-0.01,0.01)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n5_3_fold = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold))
#bf_tda_kde_5.50.5_nb.n5_3_fold
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb_n5_3_fold))
### Test set diff
#diff_drybean_tda_kde_5.50.5_nb.n5_test<-(db_svm_cf_ov_acc - nb_tda_kde_5.50.5_n5_db_nb_cf0_ov_acc)
#diff_drybean_tda_kde_5.50.5_nb.n5_test
## Bayesian Tests Test set diff
# Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n5_test<-BayesianSignTest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n5_test),-0.01,0.01)
#bst_tda_kde_5.50.5_nb.n5_test
# Odds Left Bayesian Sign Test
#bst_tda_kde_5.50.5_nb.n5_test_odds.left<-bst_tda_kde_5.50.5_nb.n5_test$probLeft/#bst_tda_kde_5.50.5_nb.n5_test$probRight
#bst_tda_kde_5.50.5_nb.n5_test_odds.left
# Bayesian Signed Rank Test
#bsr_tda_kde_5.50.5_nb.n5_test<-#BayesianSignedRank(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n5_test),-0.01,0.01)
#bsr_tda_kde_5.50.5_nb.n5_test
# Bayesian Correlated Test
#bct_tda_kde_5.50.5_nb.n5_test<-#correlatedBayesianTtest(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n5_test),0.1,-0.01,0.01)
#bct_tda_kde_5.50.5_nb.n5_test
# Rope Plot
#plot(rope(diff_drybean_tda_kde_5.50.5_nb.n5_test)))
#BayesFactor
#bf_tda_kde_5.50.5_nb.n5_test = ttestBF(x = as.matrix(diff_drybean_tda_kde_5.50.5_nb.n5_test)) #bf_tda_kde_5.50.5_nb.n5_test
#t_test
#t.test(as.matrix(diff_drybean_tda_kde_5.50.5_nb.n5_test))